Merge pull request #43 from ValentinKaisermayer/vk-more-tests

adds more tests

Author: ChrisRackauckas

src/Blocks/Blocks.jl

@@ -11,7 +11,7 @@ D = Differential(t)
 export RealInput, RealOutput, SISO
 include("utils.jl")
 
-export Gain, Sum, MatrixGain, Feedback, Add, Product, Division
+export Gain, Sum, MatrixGain, Feedback, Add, Add3, Product, Division
 export Abs, Sign, Sqrt, Sin, Cos, Tan, Asin, Acos, Atan, Atan2, Sinh, Cosh, Tanh, Exp
 export Log, Log10
 include("math.jl")

src/Blocks/continuous.jl

@@ -192,14 +192,12 @@ function PID(;name, k=1, Ti=false, Td=false, Nd=10, xi_start=0, xd_start=0)
         push!(eqs, connect(err_input, int.input))
         push!(eqs, connect(int.output, addPID.input2))
     else
-        push!(eqs, connect(err_input, Izero.input))
         push!(eqs, connect(Izero.output, addPID.input2))
     end
     if with_D
         push!(eqs, connect(err_input, der.input))
         push!(eqs, connect(der.output, addPID.input3))
     else
-        push!(eqs, connect(err_input, Dzero.input))
         push!(eqs, connect(Dzero.output, addPID.input3))
     end
     ODESystem(eqs, t, [], []; name=name, systems=sys)
@@ -285,6 +283,7 @@ function LimPID(; name, k=1, Ti=false, Td=false, wp=1, wd=1,
     )
     with_I = !isequal(Ti, false)
     with_D = !isequal(Td, false)
+    with_AWM = Ni != Inf
     if gains
         Ti = k / Ti
         Td = Td / k
@@ -302,12 +301,16 @@ function LimPID(; name, k=1, Ti=false, Td=false, wp=1, wd=1,
     @named addP = Add(k1=wp, k2=-1)
     @named gainPID = Gain(k)
     @named addPID = Add3()
+    @named limiter = Limiter(y_max=u_max, y_min=u_min)
     if with_I
-        @named addI = Add3(k1=1, k2=-1, k3=1)
+        if with_AWM
+            @named addI = Add3(k1=1, k2=-1, k3=1)
+            @named addSat = Add(k1=1, k2=-1)
+            @named gainTrack = Gain(1/(k * Ni))
+        else
+            @named addI = Add(k1=1, k2=-1)
+        end
         @named int = Integrator(k=1/Ti, x_start=xi_start)
-        @named limiter = Limiter(y_max=u_max, y_min=u_min)
-        @named addSat = Add(k1=1, k2=-1)
-        @named gainTrack = Gain(1/(k * Ni))
     else
         @named Izero = Constant(k=0)
     end
@@ -318,9 +321,12 @@ function LimPID(; name, k=1, Ti=false, Td=false, wp=1, wd=1,
         @named Dzero = Constant(k=0)
     end
 
-    sys = [reference, measurement, ctr_output, addP, gainPID, addPID]
+    sys = [reference, measurement, ctr_output, addP, gainPID, addPID, limiter]
     if with_I
-        push!(sys, [addI, int, limiter, addSat, gainTrack]...)
+        if with_AWM
+            push!(sys, [addSat, gainTrack]...)
+        end
+        push!(sys, [addI, int]...)
     else
         push!(sys, Izero)
     end
@@ -335,16 +341,18 @@ function LimPID(; name, k=1, Ti=false, Td=false, wp=1, wd=1,
         connect(measurement, addP.input2),
         connect(addP.output, addPID.input1),
         connect(addPID.output, gainPID.input),
+        connect(gainPID.output, limiter.input),
+        connect(limiter.output, ctr_output),
     ]
     if with_I
         push!(eqs, connect(reference, addI.input1))
         push!(eqs, connect(measurement, addI.input2))
-        push!(eqs, connect(gainPID.output, limiter.input))
-        push!(eqs, connect(limiter.output, ctr_output))
-        push!(eqs, connect(limiter.input, addSat.input2))
-        push!(eqs, connect(limiter.output, addSat.input1))
-        push!(eqs, connect(addSat.output, gainTrack.input))
-        push!(eqs, connect(gainTrack.output, addI.input3))
+        if with_AWM
+            push!(eqs, connect(limiter.input, addSat.input2))
+            push!(eqs, connect(limiter.output, addSat.input1))
+            push!(eqs, connect(addSat.output, gainTrack.input))
+            push!(eqs, connect(gainTrack.output, addI.input3))
+        end
         push!(eqs, connect(addI.output, int.input))
         push!(eqs, connect(int.output, addPID.input3))
     else

test/Blocks/continuous.jl

@@ -17,7 +17,8 @@ an integrator with a constant input is often used together with the system under
     sys = structural_simplify(iosys)
     prob = ODEProblem(sys, Pair[int.x=>1.0], (0.0, 1.0))
     sol = solve(prob, Rodas4())
-    @test sol[int.output.u][end] ≈ 2
+    @test all(sol[c.output.u] .≈ 1)
+    @test sol[int.output.u][end] .≈ 2 # expected solution 
 end
 
 @testset "Derivative" begin
@@ -34,7 +35,7 @@ end
     sys = structural_simplify(iosys)
     prob = ODEProblem(sys, Pair[int.x=>0.0], (0.0, 10.0))
     sol = solve(prob, Rodas4())
-    @test isapprox(sol[source.output.u], sol[int.output.u], atol=1e-1)
+    @test all(isapprox.(sol[source.output.u], sol[int.output.u], atol=1e-1))
 end
 
 @testset "PT1" begin
@@ -88,6 +89,9 @@ end
     sys = structural_simplify(model)
     prob = ODEProblem(sys, Pair[], (0.0, 100.0))
     sol = solve(prob, Rodas4())
+    # initial condition
+    @test sol[ss.x[1]][1] ≈ 0 atol=1e-3
+    @test sol[ss.x[2]][1] ≈ 0 atol=1e-3
     # equilibrium point is at [1, 0]
     @test sol[ss.x[1]][end] ≈ 1 atol=1e-3
     @test sol[ss.x[2]][end] ≈ 0 atol=1e-3
@@ -108,7 +112,8 @@ function Plant(;name, x_start=zeros(2))
 end
 
 @testset "PI" begin
-    @named ref = Constant(; k=2)
+    re_val = 2
+    @named ref = Constant(; k=re_val)
     @named pi_controller = PI(k=1, T=1)
     @named plant = Plant()
     @named fb = Feedback()
@@ -125,11 +130,13 @@ end
     sys = structural_simplify(model)
     prob = ODEProblem(sys, Pair[], (0.0, 100.0))
     sol = solve(prob, Rodas4())
-    @test sol[ref.output.u - plant.output.u][end] ≈ 0 atol=1e-3 # zero control error after 100s
+    @test all(isapprox.(sol[ref.output.u], re_val, atol=1e-3))  # check reference
+    @test sol[plant.output.u][end] ≈ re_val atol=1e-3 # zero control error after 100s
 end
 
 @testset "PID" begin
-    @named ref = Constant(; k=2)
+    re_val = 2
+    @named ref = Constant(; k=re_val)
     @named pid_controller = PID(k=3, Ti=0.5, Td=100)
     @named plant = Plant()
     @named fb = Feedback()
@@ -146,7 +153,46 @@ end
     sys = structural_simplify(model)
     prob = ODEProblem(sys, Pair[], (0.0, 100.0))
     sol = solve(prob, Rodas4())
-    @test sol[ref.output.u - plant.output.u][end] ≈ 0 atol=1e-3 # zero control error after 100s
+    @test all(isapprox.(sol[ref.output.u], re_val, atol=1e-3))  # check reference
+    @test sol[plant.output.u][end] ≈ re_val atol=1e-3 # zero control error after 100s
+
+    @testset "PI" begin
+        @named pid_controller = PID(k=3, Ti=0.5, Td=false)
+        @named model = ODESystem(
+            [
+                connect(ref.output, fb.input1), 
+                connect(plant.output, fb.input2),
+                connect(fb.output, pid_controller.err_input), 
+                connect(pid_controller.ctr_output, plant.input), 
+            ], 
+            t, 
+            systems=[pid_controller, plant, ref, fb]
+        )
+        sys = structural_simplify(model)
+        prob = ODEProblem(sys, Pair[], (0.0, 100.0))
+        sol = solve(prob, Rodas4())
+        @test all(isapprox.(sol[ref.output.u], re_val, atol=1e-3))  # check reference
+        @test sol[plant.output.u][end] ≈ re_val atol=1e-3 # zero control error after 100s
+    end
+
+    @testset "PD" begin
+        @named pid_controller = PID(k=10, Ti=false, Td=1)
+        @named model = ODESystem(
+            [
+                connect(ref.output, fb.input1), 
+                connect(plant.output, fb.input2),
+                connect(fb.output, pid_controller.err_input), 
+                connect(pid_controller.ctr_output, plant.input), 
+            ], 
+            t, 
+            systems=[pid_controller, plant, ref, fb]
+        )
+        sys = structural_simplify(model)
+        prob = ODEProblem(sys, Pair[], (0.0, 100.0))
+        sol = solve(prob, Rodas4())
+        @test all(isapprox.(sol[ref.output.u], re_val, atol=1e-3))  # check reference
+        @test sol[plant.output.u][end] > 1 # without I there will be a steady-state error
+    end
 end
 
 @test_skip begin 
@@ -236,7 +282,8 @@ end
 end
 
 @testset "LimPI" begin
-    @named ref = Constant(; k=1)
+    re_val = 1
+    @named ref = Constant(; k=re_val)
     @named pi_controller_lim = LimPI(k=3, T=0.5, u_max=1.5, u_min=-1.5, Ta=0.1)
     @named pi_controller = PI(k=3, T=0.5)
     @named sat = Limiter(y_max=1.5, y_min=-1.5)
@@ -279,15 +326,19 @@ end
         sol = solve(prob, Rodas4())
     end
 
-    @test sol[ref.output.u - plant.output.u][end] ≈ 0 atol=1e-3 # zero control error after 100s
-    @test sol_lim[ref.output.u - plant.output.u][end] ≈ 0 atol=1e-3 # zero control error after 100s
+    @test all(isapprox.(sol[ref.output.u], re_val, atol=1e-3))  # check reference
+    @test all(isapprox.(sol_lim[ref.output.u], re_val, atol=1e-3))  # check reference
+    @test sol[plant.output.u][end] ≈ re_val atol=1e-3 # zero control error after 100s
+    @test sol_lim[plant.output.u][end] ≈ re_val atol=1e-3 # zero control error after 100s
+    @test all(-1.5 .<= sol_lim[pi_controller_lim.ctr_output.u] .<= 1.5) # test limit
 
     # Plots.plot(sol; vars=[plant.output.u]) # without anti-windup measure
     # Plots.plot!(sol_lim; vars=[plant.output.u]) # with anti-windup measure
 end
 
 @testset "LimPID" begin
-    @named ref = Constant(; k=1)
+    re_val = 1
+    @named ref = Constant(; k=re_val)
     @named pid_controller = LimPID(k=3, Ti=0.5, Td=100, u_max=1.5, u_min=-1.5, Ni=0.1/0.5)
     @named plant = Plant()
     @named model = ODESystem(
@@ -304,5 +355,112 @@ end
     sol = solve(prob, Rodas4())
 
     # Plots.plot(sol, vars=[plant.output.u, plant.input.u])
-    @test sol[ref.output.u - plant.output.u][end] ≈ 0 atol=1e-3 # zero control error after 100s
+    @test all(isapprox.(sol[ref.output.u], re_val, atol=1e-3))  # check reference
+    @test sol[plant.output.u][end] ≈ re_val atol=1e-3 # zero control error after 100s
+    @test all(-1.5 .<= sol[pid_controller.ctr_output.u] .<= 1.5) # test limit
+
+    @testset "PI" begin
+        @named pid_controller = LimPID(k=3, Ti=0.5, Td=false, u_max=1.5, u_min=-1.5, Ni=0.1/0.5)
+        @named model = ODESystem(
+            [
+                connect(ref.output, pid_controller.reference), 
+                connect(plant.output, pid_controller.measurement),
+                connect(pid_controller.ctr_output, plant.input), 
+            ], 
+            t, 
+            systems=[pid_controller, plant, ref]
+        )
+        sys = structural_simplify(model)
+        prob = ODEProblem(sys, Pair[], (0.0, 100.0))
+        sol = solve(prob, Rodas4())
+
+        # Plots.plot(sol, vars=[plant.output.u, plant.input.u])
+        @test all(isapprox.(sol[ref.output.u], re_val, atol=1e-3))  # check reference
+        @test sol[plant.output.u][end] ≈ re_val atol=1e-3 # zero control error after 100s
+        @test all(-1.5 .<= sol[pid_controller.ctr_output.u] .<= 1.5) # test limit
+    end
+    @testset "PD" begin
+        @named pid_controller = LimPID(k=10, Ti=false, Td=1, u_max=1.5, u_min=-1.5)
+        @named model = ODESystem(
+            [
+                connect(ref.output, pid_controller.reference), 
+                connect(plant.output, pid_controller.measurement),
+                connect(pid_controller.ctr_output, plant.input), 
+            ], 
+            t, 
+            systems=[pid_controller, plant, ref]
+        )
+        sys = structural_simplify(model)
+        prob = ODEProblem(sys, Pair[], (0.0, 100.0))
+        sol = solve(prob, Rodas4())
+
+        # Plots.plot(sol, vars=[plant.output.u, plant.input.u])
+        @test all(isapprox.(sol[ref.output.u], re_val, atol=1e-3))  # check reference
+        @test sol[plant.output.u][end] > 0.5 # without I there will be a steady-state error
+        @test all(-1.5 .<= sol[pid_controller.ctr_output.u] .<= 1.5) # test limit
+    end
+    @testset "set-point weights" begin
+        @testset "wp" begin
+            @named pid_controller = LimPID(k=3, Ti=0.5, Td=100, u_max=1.5, u_min=-1.5, Ni=0.1/0.5, wp=0, wd=1)
+            @named model = ODESystem(
+                [
+                    connect(ref.output, pid_controller.reference), 
+                    connect(plant.output, pid_controller.measurement),
+                    connect(pid_controller.ctr_output, plant.input), 
+                ], 
+                t, 
+                systems=[pid_controller, plant, ref]
+            )
+            sys = structural_simplify(model)
+            prob = ODEProblem(sys, Pair[], (0.0, 100.0))
+            sol = solve(prob, Rodas4())
+
+            # Plots.plot(sol, vars=[plant.output.u, plant.input.u])
+            @test all(isapprox.(sol[ref.output.u], re_val, atol=1e-3))  # check reference
+            sol[pid_controller.addP.output.u] == -sol[pid_controller.measurement.u]
+            @test sol[plant.output.u][end] ≈ re_val atol=1e-3 # zero control error after 100s
+            @test all(-1.5 .<= sol[pid_controller.ctr_output.u] .<= 1.5) # test limit
+        end
+        @testset "wd" begin
+            @named pid_controller = LimPID(k=3, Ti=0.5, Td=100, u_max=1.5, u_min=-1.5, Ni=0.1/0.5, wp=1, wd=0)
+            @named model = ODESystem(
+                [
+                    connect(ref.output, pid_controller.reference), 
+                    connect(plant.output, pid_controller.measurement),
+                    connect(pid_controller.ctr_output, plant.input), 
+                ], 
+                t, 
+                systems=[pid_controller, plant, ref]
+            )
+            sys = structural_simplify(model)
+            prob = ODEProblem(sys, Pair[], (0.0, 100.0))
+            sol = solve(prob, Rodas4())
+
+            # Plots.plot(sol, vars=[plant.output.u, plant.input.u])
+            @test all(isapprox.(sol[ref.output.u], re_val, atol=1e-3))  # check reference
+            @test sol[plant.output.u][end] ≈ re_val atol=1e-3 # zero control error after 100s
+            sol[pid_controller.addD.output.u] == -sol[pid_controller.measurement.u]
+            @test all(-1.5 .<= sol[pid_controller.ctr_output.u] .<= 1.5) # test limit
+        end
+    end
+    @testset "PI without AWM" begin
+        @named pid_controller = LimPID(k=3, Ti=0.5, Td=false, u_max=1.5, u_min=-1.5, Ni=Inf)
+        @named model = ODESystem(
+            [
+                connect(ref.output, pid_controller.reference), 
+                connect(plant.output, pid_controller.measurement),
+                connect(pid_controller.ctr_output, plant.input), 
+            ], 
+            t, 
+            systems=[pid_controller, plant, ref]
+        )
+        sys = structural_simplify(model)
+        prob = ODEProblem(sys, Pair[], (0.0, 100.0))
+        sol = solve(prob, Rodas4())
+
+        # Plots.plot(sol, vars=[plant.output.u, plant.input.u])
+        @test all(isapprox.(sol[ref.output.u], re_val, atol=1e-3))  # check reference
+        @test sol[plant.output.u][end] ≈ re_val atol=1e-3 # zero control error after 100s
+        @test all(-1.5 .<= sol[pid_controller.ctr_output.u] .<= 1.5) # test limit
+    end
 end