CN113885609A - Box body temperature control method and device of vehicle-mounted refrigerator and vehicle-mounted refrigerator - Google Patents
Box body temperature control method and device of vehicle-mounted refrigerator and vehicle-mounted refrigerator Download PDFInfo
- Publication number
- CN113885609A CN113885609A CN202111239207.0A CN202111239207A CN113885609A CN 113885609 A CN113885609 A CN 113885609A CN 202111239207 A CN202111239207 A CN 202111239207A CN 113885609 A CN113885609 A CN 113885609A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- algorithm
- neural network
- box body
- temperature
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 39
- 230000036760 body temperature Effects 0.000 title claims abstract description 29
- 238000013528 artificial neural network Methods 0.000 claims abstract description 80
- 230000006870 function Effects 0.000 claims description 45
- 239000004065 semiconductor Substances 0.000 claims description 36
- 238000010438 heat treatment Methods 0.000 claims description 26
- 239000002245 particle Substances 0.000 claims description 22
- 230000000739 chaotic effect Effects 0.000 claims description 20
- 230000017525 heat dissipation Effects 0.000 claims description 20
- 238000001816 cooling Methods 0.000 claims description 17
- 238000005057 refrigeration Methods 0.000 claims description 17
- 238000009434 installation Methods 0.000 claims description 16
- 238000005457 optimization Methods 0.000 claims description 14
- 230000002068 genetic effect Effects 0.000 claims description 13
- 230000001681 protective effect Effects 0.000 claims description 12
- 238000012546 transfer Methods 0.000 claims description 11
- 230000035772 mutation Effects 0.000 claims description 10
- 238000005192 partition Methods 0.000 claims description 8
- 238000010606 normalization Methods 0.000 claims description 7
- 239000013598 vector Substances 0.000 claims description 7
- 238000004321 preservation Methods 0.000 claims description 6
- 238000010845 search algorithm Methods 0.000 claims description 6
- 230000008859 change Effects 0.000 claims description 5
- 238000011156 evaluation Methods 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 5
- 238000013209 evaluation strategy Methods 0.000 claims description 4
- 210000002569 neuron Anatomy 0.000 claims description 4
- 238000005187 foaming Methods 0.000 claims description 3
- 239000007983 Tris buffer Substances 0.000 claims description 2
- 238000007664 blowing Methods 0.000 claims description 2
- 239000004519 grease Substances 0.000 claims description 2
- 238000013507 mapping Methods 0.000 claims description 2
- 229920001296 polysiloxane Polymers 0.000 claims description 2
- 238000013139 quantization Methods 0.000 claims description 2
- 238000006467 substitution reaction Methods 0.000 claims description 2
- 238000004364 calculation method Methods 0.000 claims 1
- 230000001276 controlling effect Effects 0.000 description 12
- 238000010586 diagram Methods 0.000 description 11
- 239000000243 solution Substances 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 230000007246 mechanism Effects 0.000 description 3
- 230000033228 biological regulation Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 239000004576 sand Substances 0.000 description 2
- 230000002411 adverse Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 210000000349 chromosome Anatomy 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 230000036039 immunity Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000002028 premature Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 238000013468 resource allocation Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D23/00—Control of temperature
- G05D23/19—Control of temperature characterised by the use of electric means
- G05D23/20—Control of temperature characterised by the use of electric means with sensing elements having variation of electric or magnetic properties with change of temperature
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Feedback Control In General (AREA)
- Devices That Are Associated With Refrigeration Equipment (AREA)
Abstract
The application provides a method and a device for controlling the box body temperature of a vehicle-mounted refrigerator and the vehicle-mounted refrigerator, wherein the method comprises the following steps: determining a first parameter set of a fuzzy neural network by using the fuzzy neural network constructed for an intelligent PID controller and taking a target temperature value set for a vehicle-mounted refrigerator and an actual temperature value of a box body as input in an off-line mode; taking the determined first parameter set as an initial value of a back propagation algorithm, carrying out fuzzy reasoning on the fuzzy neural network in an online mode, and determining an optimal second parameter set of the intelligent PID controller, wherein the optimal second parameter set comprises proportional, integral and differential control parameters for reflecting the intelligent PID controller; and triggering the intelligent PID controller to control the box body temperature of the current vehicle-mounted refrigerator by using the optimal second parameter set determined by combining an off-line algorithm and an on-line algorithm. The scheme of this application can be accurate quick constant control vehicle-mounted refrigerator's box temperature.
Description
Technical Field
The invention relates to the technical field of vehicle-mounted refrigerators, in particular to a method and a device for controlling the temperature of a refrigerator body of a vehicle-mounted refrigerator and the vehicle-mounted refrigerator.
Background
The vehicle-mounted refrigerator is a refrigerated cabinet which can be carried on an automobile. The vehicle-mounted refrigerator is a new generation of refrigeration and cold storage appliance popular in the international market in recent years. Two kinds of vehicle-mounted refrigerators mainly exist in the market, one kind is a semiconductor vehicle-mounted refrigerator, and the principle of the semiconductor vehicle-mounted refrigerator is that an electronic chip is used for refrigeration; the other type is a compressor vehicle-mounted refrigerator, the compressor is the traditional technology of the traditional refrigerator, and the refrigerating temperature is low and ranges from-18 ℃ to 10 ℃.
With the development of artificial intelligence technology, the method of controlling the temperature of the refrigerator by means of an electronic chip is more and more advanced. The development of intelligent PID (proportion, integral and differential regulation) provides a new path for parameter setting of a PID controller, such as fuzzy setting based on improved cross entropy, neural network setting based on minimum resource allocation, genetic algorithm setting based on the theory of optimal point set, artificial immunity setting, improved particle swarm setting, ant colony algorithm setting and the like, and can overcome adverse factors such as large time lag, large disturbance, strong coupling and the like.
For the existing intelligent PID technology, the real-time adjustment can be divided into two types according to whether the control parameters are adjusted in real time: an offline algorithm and an online algorithm; the off-line algorithm selects an excellent algorithm to obtain an excellent PID controller parameter of the system model, and uses the excellent PID controller parameter to perform constant temperature control, but the off-line algorithm has the defect of maladjustment which is difficult to overcome when external conditions are changed violently; the online algorithm, namely the initial value of the parameter set by experience changes towards the optimal value within the specified range according to the algorithm, so as to obtain the excellent PID controller parameter, has the characteristic that the parameter is dynamically adjusted along with the working condition, but the adjustment is slow when the initial value of the parameter set by experience greatly changes with the actual system condition.
Disclosure of Invention
The invention provides a method and a device for controlling the box body temperature of a vehicle-mounted refrigerator and the vehicle-mounted refrigerator, which are used for accurately, quickly and constantly controlling the box body temperature of the vehicle-mounted refrigerator.
To solve the above technical problem, one or more embodiments of the present specification are implemented as follows:
in a first aspect, a method for controlling a temperature of a box body of an on-vehicle refrigerator is provided, which includes:
determining a first parameter set of a fuzzy neural network by using the fuzzy neural network constructed for an intelligent PID controller and taking a target temperature value set for a vehicle-mounted refrigerator and an actual temperature value of a box body as input in an off-line mode;
taking the determined first parameter set as an initial value of a back propagation algorithm, carrying out fuzzy reasoning on the fuzzy neural network in an online mode, and determining an optimal second parameter set of the intelligent PID controller, wherein the optimal second parameter set comprises proportional, integral and differential control parameters for reflecting the intelligent PID controller;
and triggering the intelligent PID controller to control the box body temperature of the current vehicle-mounted refrigerator by using the optimal second parameter set determined by combining an off-line algorithm and an on-line algorithm.
In a second aspect, a box temperature control device for an on-vehicle refrigerator is provided, comprising:
the off-line algorithm module is used for determining a first parameter set of the fuzzy neural network by utilizing the fuzzy neural network constructed for the intelligent PID controller and taking a target temperature value set for the vehicle-mounted refrigerator and an actual temperature value of the refrigerator body as input in an off-line mode;
the online algorithm module is used for taking the determined first parameter set as an initial value of a back propagation algorithm, carrying out fuzzy reasoning on the fuzzy neural network in an online mode, and determining an optimal second parameter set of the intelligent PID controller, wherein the optimal second parameter set comprises proportional, integral and differential control parameters for reflecting the intelligent PID controller;
and the PID control module triggers the intelligent PID controller to control the temperature of the box body of the current vehicle-mounted refrigerator by using the optimal second parameter set determined based on the combination of an off-line algorithm and an on-line algorithm.
In a third aspect, an in-vehicle refrigerator is provided, including: the box body, the semiconductor refrigeration piece, the inner radiating fin, the outer radiating fin, the inner circulating fan, the outer radiating fan, the temperature sensor, the control module, the partition plate and the protective cover; the control module is the box body temperature control device of the vehicle-mounted refrigerator in the second aspect, or the control module comprises the box body temperature control device of the vehicle-mounted refrigerator in the second aspect;
the box body is a heat preservation box body, and each surface forming the box body is provided with a foaming heat preservation layer; arranging a partition board in the box body, dividing the inner space of the box body into an installation chamber and a storage chamber by the partition board, arranging an air outlet and an air return opening on the partition board, arranging an installation window communicated with the installation chamber on one side wall of the box body, arranging a protective cover on the outer side of the box body, and arranging a heat dissipation air opening on the protective cover;
the semiconductor refrigerating sheet is arranged on the mounting window of the box body; the inner radiating fins are arranged in the mounting chamber and are in heat transfer contact with one side surface of the semiconductor refrigerating sheet, and the outer radiating sheets are arranged on the outer side of the box body and are in heat transfer contact with the other side surface of the semiconductor refrigerating sheet; the internal circulation fan is arranged in the installation chamber, faces the air outlet and is used for blowing low-temperature/high-temperature air flow in the installation chamber into the storage chamber; the outer heat dissipation fan is arranged on the outer side of the box body, faces the outer heat dissipation sheet, is arranged in the protective cover and is opposite to the heat dissipation air inlet, and is used for providing cooling/back-heating air flow for the outer heat dissipation sheet;
the temperature sensor is arranged in the box body, is arranged in the installation chamber at a position opposite to the air return opening, is used for acquiring the temperature of the box body, and is in signal connection with the control module; the control module is in control connection with the semiconductor refrigeration sheet, the internal circulation fan and the external heat dissipation fan in a wired connection mode; the control module, the outer heat dissipation sheet and the outer heat dissipation fan are all arranged in the protective cover.
According to the technical scheme, a fuzzy neural network constructed for an intelligent PID controller is utilized, and a first parameter set of the fuzzy neural network is determined in an off-line mode based on a target temperature value set for a vehicle-mounted refrigerator and an actual temperature value of a box body; taking the determined first parameter set as an initial value of a back propagation algorithm, carrying out fuzzy reasoning on the fuzzy neural network in an online mode, and determining an optimal second parameter set of the intelligent PID controller, wherein the optimal second parameter set comprises proportional, integral and differential control parameters for reflecting the intelligent PID controller; and triggering the intelligent PID controller to control the box body temperature of the current vehicle-mounted refrigerator by using the optimal second parameter set determined by combining the offline algorithm and the online algorithm, so that the box body temperature of the vehicle-mounted refrigerator can be accurately and quickly and constantly controlled.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
Fig. 1 is a schematic step diagram of a method for controlling a temperature of a cabinet of a vehicle-mounted refrigerator according to an embodiment of the present invention.
Fig. 2 is a schematic structural diagram of a fuzzy neural network-based PID controller according to an embodiment of the present invention.
Fig. 3 is a schematic diagram of a fuzzy neural network based on a Mamdani model according to an embodiment of the present invention.
Fig. 4 is a flowchart of a method for controlling a constant temperature of a cabinet of a vehicle-mounted refrigerator according to an embodiment of the present invention.
Fig. 5 is a schematic structural diagram of a box body temperature control device of a vehicle-mounted refrigerator according to an embodiment of the invention.
Fig. 6 is a schematic diagram of a simple structure of an in-vehicle refrigerator according to an embodiment of the present invention.
Fig. 7 is a flowchart illustrating a case temperature control rule of the in-vehicle refrigerator according to an embodiment of the present invention.
Fig. 8 is a schematic diagram of a hardware structure of a box temperature control device of a vehicle-mounted refrigerator according to an embodiment of the present invention.
Detailed Description
In order to make the technical solutions in the present specification better understood, the technical solutions in one or more embodiments of the present specification will be clearly and completely described below with reference to the accompanying drawings in one or more embodiments of the present specification, and it is obvious that the one or more embodiments described are only a part of the embodiments of the present specification, and not all embodiments. All other embodiments that can be derived by a person skilled in the art from one or more of the embodiments described herein without making any inventive step shall fall within the scope of protection of this document.
In view of the technical problems in the background art, the application provides an improved PID parameter self-tuning algorithm based on a fuzzy neural network, which specifically comprises the following steps: the fuzzy neural network based on the Mamdani model is adopted to carry out the self-adaptive tuning of the PID controller, the fuzzy neural network parameters are learned and adjusted in a particle swarm-genetic-chaotic algorithm offline coarse tuning and Back Propagation (BP) algorithm online fine tuning mode, therefore, the optimized PID control parameters are determined for the improved intelligent PID controller, the PID controller is touched to control the box body temperature of the vehicle-mounted refrigerator based on the PID control parameters, and the accurate and rapid constant control is realized.
The off-line algorithm mainly adopts an optimization algorithm combining particle swarm, heredity and chaos, namely a particle swarm-heredity-chaos algorithm, which is based on a genetic algorithm, maintains population diversity based on different individual evaluation strategies, replaces a mutation operator with a speed operator of the particle swarm algorithm, and introduces a chaos search algorithm into the genetic algorithm to further optimize individuals, and has the characteristics of rapid convergence, premature prevention and overlarge control quantity.
The online algorithm adopts a Back Propagation (BP) algorithm, and has the characteristics of higher dynamic characteristic, better reasoning capability and capability of quickly and accurately adjusting the parameters (namely the subsequently mentioned second parameter set) of the intelligent PID controller in real time.
Example one
Referring to fig. 1, a schematic step diagram of a method for controlling a box temperature of an in-vehicle refrigerator provided in an embodiment of the present specification, the method may include the following steps:
step 102: the method comprises the steps of utilizing a fuzzy neural network constructed for an intelligent PID controller, and determining a first parameter set of the fuzzy neural network in an off-line mode based on a target temperature value set for a vehicle-mounted refrigerator and an actual temperature value of a box body.
Optionally, when the step 102 determines the first parameter set of the fuzzy neural network in an offline manner based on the target temperature value set for the vehicle-mounted refrigerator and the actual temperature value of the refrigerator body as inputs by using the fuzzy neural network constructed for the intelligent PID controller, the method may specifically include:
constructing a fuzzy neural network for the intelligent PID controller; initializing a first set of parameters of a fuzzy neural network, wherein the first set of parameters comprises: network weight, membership function base width and membership function central value; using a genetic algorithm to carry out initial value coding on each parameter in the first parameter set, and calculating the individual fitness value of the population; selecting by adopting a roulette method, and selecting high-quality individuals to be inherited to the next generation; adopting a self-adaptive single-point crossing operator to cross every two new population individuals obtained after the selection operation to obtain a new generation population; adopting a speed operator to carry out individual updating on the new generation population obtained after the cross operation, generating a new population again, and calculating the individual fitness value of the newly generated population; performing chaotic search optimization on the population, judging whether the evolution algebra G is terminated, and if the judgment result is negative, returning to execute again: calculating individual fitness value of the population; and if so, acquiring the optimal first parameter set.
On the basis of the original PID controller, a fuzzy neural network is utilized, a group of suitable PID parameters are set by taking a target temperature value and an actual box body temperature value of a system as input, and the PID controller outputs a control signal to control a controlled object, namely a semiconductor refrigerating sheet. And obtaining an approximate optimal solution of each parameter of the fuzzy neural network by utilizing the particle swarm-genetic-chaotic algorithm offline optimization. Through the learning function of the neural network, the output layer weight of the network, the central value and the width of the Gaussian membership function are adjusted on line, and therefore the self-adaptive control of the PID control parameters is achieved.
Step 104: and taking the determined first parameter set as an initial value of a back propagation algorithm, carrying out fuzzy reasoning on the fuzzy neural network in an online mode, and determining an optimal second parameter set of the intelligent PID controller, wherein the optimal second parameter set comprises proportional, integral and differential control parameters for reflecting the intelligent PID controller.
The technical scheme adopts an optimization algorithm combining particle swarm, heredity and chaos which mainly adopts a genetic algorithm: the particle swarm-genetic-chaotic algorithm adopts different individual evaluation strategies to keep population diversity, replaces a mutation operator with a speed operator of the particle swarm algorithm, and introduces a chaotic search algorithm into the genetic algorithm to further optimize individuals.
Step 106: and triggering the intelligent PID controller to control the box body temperature of the current vehicle-mounted refrigerator by using the optimal second parameter set determined by combining an off-line algorithm and an on-line algorithm.
In the invention, the box body temperature control method is specifically designed by the following steps:
-designing a model of a temperature control system
According to the thermodynamic theory, the heat generated by the semiconductor refrigeration sheet during operation is as follows:
wherein E is the heating coefficient of the semiconductor refrigerating sheet; t is0Is the initial temperature; t is1The temperature value of the box body is the temperature value of the refrigerating sheet after working for unit time; k is the heat transfer coefficient of the semiconductor refrigerating sheet; a is the contact area of the heat conducting sheets at the two ends of the semiconductor refrigerating sheet; and deltaQ is the heat generated by the semiconductor refrigerating sheet in unit time.
The heat Q generated per unit time according to the law of conservation of energy is:
according to the formula (2), the heat Q and the voltage U across the semiconductor chilling plate are in a nonlinear relationship, and the semiconductor chilling plate is linearized according to the thermal inertia thereof by:
finally, formula (3) is taken into formula (1), and Δ T is made equal to T1-T0And obtaining a semiconductor chilling plate model:
and (3) performing Laplace transformation on the formula (4) to obtain a transfer function of the semiconductor refrigerating sheet:
wherein ,
considering the time lag of the temperature control system, the loop of transferring heat in the box body is represented by a first-order inertia element with delay:
finally, obtaining a temperature control system model for offline optimization of the particle swarm-genetic-chaotic algorithm:
in the invention, the temperature values of the system in the processes of temperature rise and temperature drop are recorded, MATLAB is used for fitting, a transfer function model is built, and finally the parameters K, T are determined by observing a transfer function curve and a temperature-time curve1、T2τ. It should be understood that the transfer coefficient of the semiconductor chilling plates determined in the design step is the above-mentioned final evolution algebra G.
Referring to fig. 2, a schematic structural diagram of a PID controller based on a fuzzy neural network constructed in an embodiment of the present specification; the temperature control model constructed by the fuzzy neural network can determine optimal parameters in an off-line and on-line combined mode, adaptively adjust and determine proportional, integral and differential control parameters of the intelligent PID controller based on the optimal parameters, and feed back the parameters to the intelligent PID controller to perform constant temperature control on receptor objects such as a box body of the vehicle-mounted refrigerator.
-designing an improved PID parameter self-tuning algorithm based on a fuzzy neural network
Based on the original PID controller, the invention uses the fuzzy neural network, uses the target temperature value and the actual box temperature value of the system as input, sets a group of suitable PID parameters, and the PID controller outputs a control signal to control the controlled object, namely the semiconductor refrigerating sheet. And obtaining an approximate optimal solution of each parameter of the fuzzy neural network by utilizing the particle swarm-genetic-chaotic algorithm offline optimization. Through the learning function of the neural network, the output layer weight of the network, the central value and the width of the Gaussian membership function are adjusted on line, and therefore the self-adaptive control of the PID control parameters is achieved. The improved PID parameter self-tuning algorithm based on the fuzzy neural network is divided into the following modules:
(1) PID controller module based on fuzzy neural network
In the invention, a target temperature value and an actual box body temperature value of a system are used as input vectors of a fuzzy neural network, and the quantization grade of the input vectors is defined as 7, namely a fuzzy division number mi7(i ═ 1,2), which is described in fuzzy language: { NB, NM, NS, Z0, PS, PM, PB }.
In the invention, the temperature deviation e (n) and the temperature change rate ec (n) are used as input variables of the PID controller.
wherein ,
t is the time; n is the number of refrigeration/heating control cycles, TintervalAs algorithm parametersAdjusting the output time interval of the refrigerating sheet; r isin(n) is the desired output of the system, i.e., the target temperature value for the nth cooling/heating control cycle, youtAnd (n) is the actual output of the system, namely the box temperature value of the nth cooling/heating control period.
In the invention, the control quantity of the incremental PID controller is as follows:
in the invention, the fuzzy neural network based on the Mamdani model is a 5-layer feedforward network which is respectively an input layer, a fuzzy layer, an inference layer, a normalization layer and an output layer, and the topological structure of the fuzzy neural network is as follows: a → 2B → B2→B2→ C. As shown in FIG. 3, the signal is 2 → 14 → 49 → 49 → 3 in the present invention. Wherein x isiRepresenting an input, is an input vector X ═ X1,x2,…,xn]TComponent of (a), ωijThe connection weight from the normalization layer to the output layer is 1, and the weights between nodes in each layer are not shown in fig. 2.
An input layer: each node of the layer is directly connected with an input vector X ═ X1,x2,…,xg]TThe components of (a) are connected.
f1(i,j)＝xij＝xi (10)
Wherein, i is the serial number of the number of input variables; j is a fuzzy division number.
Blurring layer: each neuron node of the layer represents a fuzzy language such as NB, NM, NS, etc., and is formulated as follows using a gaussian function as an evaluation criterion.
wherein ,cij、σijThe center, width, x of the membership function of the jth fuzzy set of the ith input variableiIs the ith input variable, i ═1,2,…,g；j＝1,2,…mi。
A fuzzy inference layer: each neuron node of the layer corresponds to each fuzzy rule of the fuzzy rule library, and each fuzzy rule is paired to calculate the fitness of each node.
wherein ,
a normalization layer: the layer performs network structure overall normalization operation.
Wherein l is 1,2, …, m.
An output layer: this layer implements the anti-fuzzy computation, i.e. the output is Kp、Ki、KdAnd (4) setting results.
Wherein k is 1,2, …, r, r is the number of output parameters.
In the present invention, r is 1,2,3,. Kp、Ki、KdThe following were used:
Kp＝f5(1) (15)
Ki＝f5(2) (16)
Kd＝f5(3) (17)。
(2) off-line algorithm module
In the technical scheme, an optimization algorithm combining particle swarm, heredity and chaos, namely a particle swarm-heredity-chaos algorithm, which is based on a genetic algorithm is adopted, and different individual evaluation strategies are adopted to maintain the seedsAnd (3) group diversity, replacing a mutation operator with a speed operator of a particle swarm algorithm, and introducing a chaotic search algorithm into the genetic algorithm to further optimize individuals. The particle swarm-genetic-chaotic algorithm respectively uses the length of
The binary code string of (a) represents each decision variable. By using the technical thought of the niche, each generation of individuals is divided into 2 generations of individuals according to the order s of the initial species group by adopting an even segmentation strategy based on individual codingsAnd (4) sub-populations, wherein the rightmost s binary numbers of individuals in each sub-population are the same. Simultaneously generating the optimal number of individuals T before the a generation by using the sub-population jjaThe total optimal individual number T generated before the a generation compared with the whole populationaRatio of (delta)aAnd dynamically regulating the evolution state of each sub-population.
wherein ,
the niche technology is that each generation of individuals is divided into several classes, several individuals with high fitness in each class are selected as excellent representatives of one class to form a group, and then the individuals are hybridized and mutated in the group and among different groups to generate a new generation of individual group. And simultaneously, a pre-selection mechanism and a displacement mechanism or a sharing mechanism are adopted to complete the task. Based on the Genetic algorithm (NGA) of the niche, the diversity of solutions can be better kept, meanwhile, the global optimization capability and the convergence rate are high, and the method is particularly suitable for the optimization problem of complex multi-peak functions.
In order to obtain satisfactory dynamic characteristics of the transition process and prevent overlarge control quantity, J (x) is used for evaluating the individual fitness:
ey(t)＝yout(t)-yout(t-1) (20)
e(t)＝rin(t)-yout(t) (21)
wherein ,rin(t) target temperature value, which is the expected output of the fuzzy neural network at the tth moment; y isout(t) the actual output of the fuzzy neural network at the tth moment is the box temperature value; e (t) is the error of the fuzzy neural network at the t moment; u (t) is the PID controller control quantity at the t-th moment; t is trIs the rise time; ey (t) is the overshoot of the fuzzy neural network at the t-th moment.
Selecting the reciprocal of the individual fitness evaluation function J (x) as an objective function, and selecting the following formula (23) as the individual fitness function in order to ensure that a small number of individuals with extremely high adaptability in the population are not eliminated prematurely:
M(xi)＝F(xi)+F(xi)/N(i) (23)
wherein ,F(xi) (i) is the ratio of the objective function itself of the individual i to the number of times the individual is selected; n (i) is the number of times the sub-population is selected.
The selection operation adopts a roulette method, and the probability that the individual i is selected to be inherited to the next generation group is as follows:
wherein N is the population size, M (x)i) The fitness of the individual i.
The following single-point crossover operators are used in the crossover operation, and pairwise crossover is carried out on individuals in the selected population to obtain a new generation of population.
wherein ,fmaxIs the maximum fitness value in the population; f. ofavgMean fitness value for each generation population; f' is the greater fitness value of the two individuals to be crossed; pc1、Pc2Respectively, the maximum and minimum of the cross probability.
And replacing a mutation operator with a speed operator of the particle swarm algorithm, so as to realize individual updating and regenerate a new population.
δa＝1-ρa (26)
Xi(t+1)＝Xi(t)+Vi(t+1) (28)
wherein ,ρaThe variation probability of the sub-population j generation a is used for replacing the inertia weight of the particle swarm algorithm; rhojThe variation probability of the t generation of the sub-population j; r is1、r2Are respectively the interval [0,1]A random number within; c. C1、c2Represents a learning factor;
the current optimal position of the individual i;
is the current optimal position of the sub-population j. The formula (26) is as follows: the more excellent individuals are generated in the population → the ratio deltaaThe greater → the probability of variation ρaSmaller → more stable population, and fewer individuals superior in population generation → the ratio deltaaSmaller → the variation probability ρaThe larger → the more prone to mutation the population. Thus using the probability of variation ρaThe inertia weight of the particle swarm algorithm is replaced, so that a better effect can be achieved.
In the chaotic search algorithm, the change range of chaotic variables is respectively changed to the value range of corresponding optimized variables, and the classical Logistic mapping is adopted to generate chaotic motion, wherein the chaotic equation is shown in the following formula. Setting the individuals in the newly generated population, namely each chromosome, as the current optimal value, performing chaotic optimization search on the individuals with higher fitness values in the population, calculating the fitness value of a feasible solution experienced by each chaotic variable, and keeping the solution with the best performance until the maximum generation number of the chaotic search is reached, if the fitness value of the generated new individual is greater than that of the original individual, replacing the original individual, otherwise, keeping the original individual unchanged.
zi(k+1)＝μzi(k)[1-zi(k)] (29)
k＝0,1,…,N
wherein ,zi(k) The value of the ith chaotic variable after the kth chaotic iteration is obtained; mu is a control parameter, and mu is 4, namely, the device is completely in a chaotic state.
(3) Online algorithm module
In the invention, the BP algorithm selects a mean square error loss function as a performance index function:
where e (n) is the control error for each iteration step. By a symbolic function
Approximate substitution
The error is compensated by adjusting learning rate eta, and the calculated network weight omegaijMembership function base width bijCenter value of membership function cijComprises the following steps:
wherein the learning rate eta is eta1＝η2＝η3> 0, alpha is a momentum factor.
The following describes a method for controlling the constant temperature of the refrigerator body of the vehicle-mounted refrigerator through a specific control example, as shown in fig. 4, the method specifically includes the following steps:
a1: constructing a fuzzy neural network and a PID controller: make fuzzy division number mi7(i ═ 1,2), the topology is 2 → 14 → 49 → 49 → 3.
A2: initializing relevant parameters: e.g. code string length
The final evolution algebra G is 100, the linear coefficient k is 100, Pc1＝0.9，Pc20.6, 20 chaotic iteration times N, 0.2 learning rate eta, 0.02 momentum factor alpha and algorithm parameter regulation refrigerating sheet output time interval TintervalThe BP algorithm online fine-adjusts the maximum duration T as 3smax＝5min。
A3: initial fuzzy neural network omegaij、mij、cijLength: such as network weight omegaijThe value interval is [ -1, +1]Membership function base width bijThe value interval is [0.1, + 3%]Center value of membership function cijThe value interval is [ -3, +3]。
A4: using genetic algorithm to pair omegaij、bij、cijThe initial value is encoded, and equation (23) is used as the individual fitness function.
A5: and calculating individual fitness values of the population.
A6: selecting by roulette formula (24), selecting high-quality individuals and inheriting to the next generation.
A7: and (3) crossing the new population individuals obtained after the selection operation pairwise by adopting a self-adaptive single-point crossing operator (25) to obtain a new generation population.
A8: and (3) carrying out individual updating on the new generation population obtained after the cross operation by adopting a speed operator (26-28) to generate a new population again.
A9: and calculating the individual fitness value of the newly generated population.
A10: and performing chaotic search optimization on the population.
A11: judging whether the evolution algebra G is terminated or not, if not, returning to execute A5, and recalculating the fitness value; if the judgment result is yes, finishing the A4-A11 offline particle swarm-genetic-chaotic algorithm module to obtain the optimal omegaij、bij、cij。
A12: obtaining the optimal omega by an off-line algorithm after the vehicle-mounted refrigerator is powered on every timeij、bij、cijAs an initial value for online fine tuning of the BP algorithm.
A13: fuzzy inference is carried out by using a fuzzy neural network.
A14: the error is calculated using the mean square error loss function (30) as the performance index function.
A15: calculating the local gradient of each layer, and further updating the fuzzy neural network parameter omegaij、bij、cij。
A16: using updated parameter omegaij、bij、cijFinding updated Kp、Ki、Kd。
A17: judging whether the BP algorithm on-line fine adjustment reaches the maximum time length TmaxIf the judgment result is negative, returning to execute A13, and performing fuzzy processing again; if the judgment result is yes, finishing the fine adjustment of the A12-A17 online BP algorithm to obtain the optimal Kp、Ki、KdAnd using the value to perform subsequent constant temperature control.
Further, in a specific thermostatic control process, a target temperature r of a corresponding mode set for a cabinet system of the in-vehicle refrigerator is readin(n)。
Detection vehicleCurrent cabinet temperature y of refrigeratorout(t) and let yout(n)＝TnWherein n is the number of refrigeration/heating control cycles;
comparing the current box temperature TnAnd the target temperature TsThe difference between the temperature difference and the artificially set temperature variation delta T,
in the cooling mode: when T isn≥(Ts_c+ Δ T), working according to the maximum refrigeration capacity; when | Tn-Ts_cWhen | < delta T, using the optimal second parameter set self-tuning algorithm of the improved intelligent PID controller based on the fuzzy neural network to enable TnQuickly reach Ts_cAnd remain unchanged; when T isn≤(Ts_cAt) then stop output, wait for temperature return, when Tn≥(Ts_c+ delta T), starting to work according to the maximum refrigerating capacity;
in the heating mode: when T isn≤(Ts_h- Δ T), operating at maximum heating capacity; when | Tn-Ts_hWhen | < delta T, using the optimal second parameter set self-tuning algorithm of the improved intelligent PID controller based on the fuzzy neural network to enable TnQuickly reach Ts_hAnd remain unchanged; when T isn≤(Ts_h+ Δ T), stop output, wait for cooling, when Tn≤(Ts_h- Δ T), operation according to the maximum heating capacity is resumed;
and working according to the determined actual output refrigerating/heating quantity of the semiconductor refrigerating sheet in the current control period, and returning to repeatedly execute control when the next period is up.
According to the technical scheme, a fuzzy neural network constructed for an intelligent PID controller is utilized, and a first parameter set of the fuzzy neural network is determined in an off-line mode based on a target temperature value set for a vehicle-mounted refrigerator and an actual temperature value of a refrigerator body; taking the determined first parameter set as an initial value of a back propagation algorithm, carrying out fuzzy reasoning on the fuzzy neural network in an online mode, and determining an optimal second parameter set of the intelligent PID controller, wherein the optimal second parameter set comprises proportional, integral and differential control parameters for reflecting the intelligent PID controller; and triggering the intelligent PID controller to control the box body temperature of the current vehicle-mounted refrigerator by using the optimal second parameter set determined by combining an off-line algorithm and an on-line algorithm. The scheme of this application can be accurate quick constant control vehicle-mounted refrigerator's box temperature.
Example two
Referring to fig. 5, a schematic structural diagram of a box temperature control device of an in-vehicle refrigerator provided in an embodiment of the present specification includes:
an offline algorithm module 502, which determines a first parameter set of a fuzzy neural network in an offline manner based on a target temperature value set for the vehicle-mounted refrigerator and an actual temperature value of the refrigerator body as inputs, by using the fuzzy neural network constructed for the intelligent PID controller;
the online algorithm module 504 is configured to perform fuzzy inference on the fuzzy neural network in an online manner by using the determined first parameter set as an initial value of a back propagation algorithm, and determine an optimal second parameter set of the intelligent PID controller, where the optimal second parameter set includes proportional, integral, and differential control parameters for reflecting the intelligent PID controller;
and the PID control module 506 triggers the intelligent PID controller to control the temperature of the box body of the current vehicle-mounted refrigerator by using the optimal second parameter set determined based on the combination of an off-line algorithm and an on-line algorithm.
It is understood that other functions and effects of the box body temperature control device of the vehicle-mounted refrigerator can be referred to the method embodiment in the first embodiment.
EXAMPLE III
Referring to fig. 6, a schematic diagram of a simple structure of an in-vehicle refrigerator provided in an embodiment of the present disclosure is shown, where the in-vehicle refrigerator may include: the box body (I), the semiconductor refrigeration piece (II), interior fin (III), outer fin (IV), internal circulation fan (V), outer fan (VI), temperature sensor (VII), control module group (VIII), baffle (IX) and protection shroud (X) constitute, and concrete structure and function explanation are as follows:
the box body (I) is a heat preservation box body, and each face forming the box body (I) is provided with a foaming heat preservation layer (XI). Set up baffle (IX) in box (I), by baffle (IX) will installation room (XII) and storeroom (XIII) are cut apart into to box (I) inner space, in air outlet (XIV), return air inlet (XV) have been seted up on baffle (IX), in set up on the lateral wall of box (I) with the communicating installation window of installation room (XII), in the outside of box (I) is provided with protecting cover lid (X), in set up the heat dissipation wind gap on protecting cover lid (X).
The semiconductor refrigerating sheet (II) is arranged on the mounting window of the box body (I); the inner radiating fins (III) are arranged in the mounting chamber (XII) and are in heat transfer contact with one side surface of the semiconductor refrigerating fin (II), and the outer radiating fins (IV) are arranged on the outer side of the box body (I) and are in heat transfer contact with the other side surface of the semiconductor refrigerating fin (II); the internal circulation fan (V) is arranged in the mounting chamber (XII), faces the air outlet (XIV), and blows a low-temperature/high-temperature air flow in the mounting chamber (XII) into the storage chamber (XIII); the outer cooling fan (VI) is arranged on the outer side of the box body (I), faces the outer cooling fins (IV), is arranged in the protective cover cap (X) and is opposite to the cooling air inlet, and is used for providing cooling/back-heating air flow for the outer cooling fins (IV);
the temperature sensor (VII) is arranged in the box body (I), is arranged in the installation chamber (XII) and is opposite to the air return opening (XV), is used for acquiring the temperature of the box body (I), and is in signal connection with the control module (VIII); the control module (VIII) is in control connection with the semiconductor refrigeration sheet (II), the internal circulation fan (V) and the external heat dissipation fan (VI) in a wired connection mode; the control module (VIII), the outer heat dissipation sheet (IV) and the outer heat dissipation fan (VI) are all arranged in the protective cover cap (X).
Wherein the control module may include: a temperature control module. The temperature control module consists of the following modules: the device comprises a receiving module and a processing module; the receiving module is used for receiving a mode from external input; the processing module is used for controlling the refrigerator to be controlled to operate according to an in-refrigerator temperature control rule, wherein the in-refrigerator temperature control rule is used for defining a refrigerator working rule which changes along with the change of the in-refrigerator temperature.
The processing module comprises: and acquiring the temperature in the refrigerator, and controlling the output of the actual refrigerating/heating quantity by PWM (pulse width modulation) voltage.
The in-box temperature control rule, as shown in fig. 7, includes:
b1: reading a target temperature r of a corresponding mode set for a cabinet system of an in-vehicle refrigeratorin(n)。
B2: detecting the current box body temperature y of the vehicle-mounted refrigeratorout(t) and let yout(n)＝TnAnd n is the number of the cooling/heating control cycles.
B3: comparing the current box temperature TnAnd the target temperature TsAnd the magnitude of the artificially set temperature variation Δ T.
In the cooling mode:
when T isn≥(Ts_c+ Δ T), working according to the maximum refrigeration capacity;
when | Tn-Ts_cWhen | < delta T, using the optimal second parameter set self-tuning algorithm of the improved intelligent PID controller based on the fuzzy neural network to enable TnQuickly reach Ts_cAnd remain unchanged; when T isn≤(Ts_c- Δ T), stopping output and waiting for temperature return;
when T isn≥(Ts_c+ delta T), starting to work according to the maximum refrigerating capacity;
in the heating mode:
when T isn≤(Ts_h- Δ T), operating at maximum heating capacity;
when | Tn-Ts_hWhen | < delta T, using the optimal second parameter set self-tuning algorithm of the improved intelligent PID controller based on the fuzzy neural network to enable TnQuickly reach Ts_hAnd remain unchanged;
when T isn≤(Ts_h+ Δ T), stop output, wait for cooling, when Tn≤(Ts_hAt) and operation according to the maximum heating capacity is resumed.
B4: and working according to the determined actual output refrigerating/heating quantity of the semiconductor refrigerating chip in the current control period, and returning to B2 to repeatedly execute control when the next period is up.
Optionally, the inner heat sink and the outer heat sink are both fin type heat sinks; and/or; and heat-conducting silicone grease layers are arranged between the inner radiating fins and the semiconductor refrigerating fins and between the outer radiating fins and the semiconductor refrigerating fins.
It is to be understood that the illustrated construction of the embodiments of the invention does not constitute a specific limitation on the apparatus/devices described above. In other embodiments of the invention, the thermostat control device/arrangement described above may include more or fewer components than shown, or combine certain components, or split certain components, or a different arrangement of components. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.
Because the information interaction, execution process, and other contents between the units in the device are based on the same concept as the method embodiment of the present invention, specific contents may refer to the description in the method embodiment of the present invention, and are not described herein again.
An embodiment of the present invention further provides a device for controlling a temperature of a box body of a vehicle-mounted refrigerator, which is shown in fig. 8 and includes: at least one memory and at least one processor;
the at least one memory to store a machine readable program;
the at least one processor is used for calling the machine readable program to execute the box body temperature control method of the vehicle-mounted refrigerator in any embodiment of the invention.
The embodiment of the invention also provides a computer readable medium, wherein a computer instruction is stored on the computer readable medium, and when the computer instruction is executed by a processor, the processor is enabled to execute the box body temperature control method of the vehicle-mounted refrigerator in any embodiment of the invention. Specifically, a system or an apparatus equipped with a storage medium on which software program codes that realize the functions of any of the above-described embodiments are stored may be provided, and a computer (or a CPU or MPU) of the system or the apparatus is caused to read out and execute the program codes stored in the storage medium.
In this case, the program code itself read from the storage medium can realize the functions of any of the above-described embodiments, and thus the program code and the storage medium storing the program code constitute a part of the present invention.
Examples of the storage medium for supplying the program code include a floppy disk, a hard disk, a magneto-optical disk, an optical disk (e.g., CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD + RW), a magnetic tape, a nonvolatile memory card, and a ROM. Alternatively, the program code may be downloaded from a server computer via a communications network.
Further, it should be clear that the functions of any one of the above-described embodiments may be implemented not only by executing the program code read out by the computer, but also by causing an operating system or the like operating on the computer to perform a part or all of the actual operations based on instructions of the program code.
Further, it is to be understood that the program code read out from the storage medium is written to a memory provided in an expansion board inserted into the computer or to a memory provided in an expansion unit connected to the computer, and then causes a CPU or the like mounted on the expansion board or the expansion unit to perform part or all of the actual operations based on instructions of the program code, thereby realizing the functions of any of the above-described embodiments.
It should be noted that not all steps and modules in the above flows and system structure diagrams are necessary, and some steps or modules may be omitted according to actual needs. The execution order of the steps is not fixed and can be adjusted as required. The system structure described in the above embodiments may be a physical structure or a logical structure, that is, some modules may be implemented by the same physical entity, or some modules may be implemented by a plurality of physical entities, or some components in a plurality of independent devices may be implemented together.
In the above embodiments, the hardware unit may be implemented mechanically or electrically. For example, a hardware element may comprise permanently dedicated circuitry or logic (such as a dedicated processor, FPGA or ASIC) to perform the corresponding operations. The hardware elements may also comprise programmable logic or circuitry, such as a general purpose processor or other programmable processor, that may be temporarily configured by software to perform the corresponding operations. The specific implementation (mechanical, or dedicated permanent, or temporarily set) may be determined based on cost and time considerations.
While the invention has been shown and described in detail in the drawings and in the preferred embodiments, it is not intended to limit the invention to the embodiments disclosed, and it will be apparent to those skilled in the art that various combinations of the code auditing means in the various embodiments described above may be used to obtain further embodiments of the invention, which are also within the scope of the invention.
Claims (10)
1. A box body temperature control method of a vehicle-mounted refrigerator is characterized by comprising the following steps:
determining a first parameter set of a fuzzy neural network by using the fuzzy neural network constructed for an intelligent PID controller and taking a target temperature value set for a vehicle-mounted refrigerator and an actual temperature value of a box body as input in an off-line mode;
taking the determined first parameter set as an initial value of a back propagation algorithm, carrying out fuzzy reasoning on the fuzzy neural network in an online mode, and determining an optimal second parameter set of the intelligent PID controller, wherein the optimal second parameter set comprises proportional, integral and differential control parameters for reflecting the intelligent PID controller;
and triggering the intelligent PID controller to control the box body temperature of the current vehicle-mounted refrigerator by using the optimal second parameter set determined by combining an off-line algorithm and an on-line algorithm.
2. The method for controlling the temperature of the refrigerator body of the vehicle-mounted refrigerator according to claim 1, wherein the determining the first parameter set of the fuzzy neural network in an off-line manner based on the target temperature value set for the vehicle-mounted refrigerator and the actual temperature value of the refrigerator body as inputs by using the fuzzy neural network constructed for the intelligent PID controller specifically comprises:
constructing a fuzzy neural network for the intelligent PID controller;
initializing a first set of parameters of a fuzzy neural network, wherein the first set of parameters comprises: network weight, membership function base width and membership function central value;
using a genetic algorithm to carry out initial value coding on each parameter in the first parameter set, and calculating the individual fitness value of the population;
selecting by adopting a roulette method, and selecting high-quality individuals to be inherited to the next generation; adopting a self-adaptive single-point crossing operator to cross every two new population individuals obtained after the selection operation to obtain a new generation population;
adopting a speed operator to carry out individual updating on the new generation population obtained after the cross operation, generating a new population again, and calculating the individual fitness value of the newly generated population;
performing chaotic search optimization on the population, judging whether the evolution algebra G is terminated, and if the judgment result is negative, returning to execute again: calculating individual fitness value of the population; and if so, acquiring the optimal first parameter set.
3. The method for controlling the box body temperature of the vehicle-mounted refrigerator according to claim 2, wherein a fuzzy neural network is constructed for the intelligent PID controller, and specifically comprises the following steps:
taking a target temperature value and an actual box body temperature value of the system as input vectors of the fuzzy neural network, and defining the quantization grade of the input vectors as 7, namely a fuzzy division number mi7(i ═ 1,2), which is described in fuzzy language: { NB, NM, NS, Z0, PS, PM, PB };
taking the temperature deviation e (n) and the temperature change rate ec (n) as input variables of the PID controller;
wherein ,
t is the time; n is the number of refrigeration/heating control cycles, TintervalAdjusting the output time interval of the refrigerating sheet for the algorithm parameter; r isin(n) is the desired output of the system, i.e., the target temperature value for the nth cooling/heating control cycle, yout(n) is the actual output of the system, namely the box temperature value of the nth refrigeration/heating control cycle;
the control quantity of the incremental PID controller is as follows:
the fuzzy neural network based on the Mamdani model is a 5-layer feedforward network which is respectively an input layer, a fuzzy layer, an inference layer, a normalization layer and an output layer, and the topological structure of the fuzzy neural network is as follows: a → 2B → B2→B2→C； wherein ,
an input layer: each node of the layer is directly connected with an input vector X ═ X1,x2,…,xg]TThe components of (a) are connected;
f1(i,j)＝xij＝xi (10)
wherein, i is the serial number of the number of input variables; j is the serial number of the fuzzy division number;
blurring layer: each neuron node of the layer represents a fuzzy language such as NB, NM and NS, and a Gaussian function is used as an evaluation standard, and the formula is expressed as follows;
wherein ,ij、σijthe center, width, x of the membership function of the jth fuzzy set of the ith input variableiIs the ith input variable, i ═ 1,2, …, g; j is 1,2, … mi；
A fuzzy inference layer: each neuron node of the layer corresponds to each fuzzy rule of the fuzzy rule library, and each fuzzy rule is paired to calculate the fitness of each node;
wherein ,
a normalization layer: the layer carries out overall normalization operation of the network structure;
wherein, l is 1,2, …, m;
an output layer: the layer realizes the anti-fuzzy calculation, namely the output is the proportional, integral and differential control parameter Kp、Ki、KdThe setting result of (1);
wherein k is 1,2, …, r and r are the number of output parameters;
meanwhile, r is 1,2, 3; kp、Ki、KdThe following were used:
Kp＝f5(1) (15)
Ki＝f5(2) (16)
Kd＝f5(3) (17)。
4. the box temperature control method of the vehicle-mounted refrigerator according to claim 2, wherein when the optimization is performed by the genetic algorithm in an off-line manner, different individual evaluation strategies are adopted to maintain population diversity, a speed operator of the particle swarm algorithm is used to replace a mutation operator, and a chaotic search algorithm is introduced into the genetic algorithm to further optimize individuals; the method specifically comprises the following steps:
dividing each generation of individuals into 2 according to the order s of the initial seed group by adopting an individual coding-based uniform division strategysSub-populations, wherein the s rightmost binary numbers of individuals in each sub-population are the same; simultaneously generating the optimal number of individuals T before the a generation by using the sub-population jjaThe total optimal individual number T generated before the a generation compared with the whole populationaRatio of (delta)aDynamically adjusting the evolution state of each sub-population;
wherein ,
j9x) was used for individual fitness evaluation:
ey(t)＝yout(t)-yout(t-1) (20)
e(t)＝rin(t)-yout(t) (21)
wherein ,rin(t) the expected output of the fuzzy neural network at the tth moment, namely the target temperature value; y isout(t) actual output of the fuzzy neural network at the tth moment, namely a box temperature value; e (t) is the error of the fuzzy neural network at the t moment; u (t) is the intelligent PI at the t-th momentD, controlling quantity of the controller; t is trIs the rise time; ey (t) is the overshoot of the fuzzy neural network at the t moment;
selecting the reciprocal of the individual fitness evaluation function J (x) as an objective function, and selecting the following formula (23) as the individual fitness function:
M(xi)＝F(xi)+F(xi)/N(i) (23)
wherein ,F(xi) (i) is the ratio of the objective function itself of the individual i to the number of times the individual is selected; n (i) the number of times the child was selected;
the selection operation adopts a roulette method, and the probability that the individual i is selected to be inherited to the next generation group is as follows:
wherein N is the population size, M (x)i) Fitness of the individual i;
the following single-point crossover operators are used in the crossover operation, and pairwise crossover is carried out on the individuals in the selected population to obtain a new generation of population;
wherein ,fmaxIs the maximum fitness value in the population; f. ofaνgMean fitness value for each generation population; f' is the greater fitness value of the two individuals to be crossed; pc1、Pc2Respectively the maximum value and the minimum value of the cross probability;
and replacing a mutation operator with a speed operator of the particle swarm algorithm, so as to realize individual updating and regenerate a new population.
δa＝1-ρa (26)
Xi(t+1)＝Xi(t)+Vi(t+1) (28)
wherein ,ρaThe variation probability of the sub-population j generation a is used for replacing the inertia weight of the particle swarm algorithm; rhojThe variation probability of the t generation of the sub-population j; r is1、r2Are respectively the interval [0,1]A random number within; c. C1、c2Represents a learning factor;
the current optimal position of the individual i;
the current optimal position of the sub-population j is obtained; the formula (26) is as follows: the more excellent individuals are generated in the population, the ratio deltaaThe greater the mutation probability ρaThe smaller the size, the more stable the heteropopulation, and the fewer individuals superior in population generation, the ratio deltaaThe smaller the mutation probability ρaThe larger, the more prone the population to mutation;
in the chaos search algorithm, the change range of the chaos variable is respectively changed to the value range of the corresponding optimization variable, and the classical Logistic mapping is adopted to generate chaos motion, and the chaos equation is shown as the following formula:
wherein ,zi(k) The value of the ith chaotic variable after the kth chaotic iteration is obtained; mu is a control parameter, and mu is 4, namely, the device is completely in a chaotic state.
5. The method for controlling the box temperature of the vehicle-mounted refrigerator according to claim 1, wherein the step of performing fuzzy inference on the fuzzy neural network in an online manner by using the determined first parameter set as an initial value of a back propagation algorithm to determine an optimal second parameter set of the intelligent PID controller specifically comprises the steps of:
after the vehicle-mounted refrigerator is powered on every time, a first parameter set obtained in an off-line mode is used as an initial value of online fine adjustment of a back propagation algorithm, and fuzzy reasoning is carried out by using a fuzzy neural network;
calculating errors by taking the following mean square error loss function as a performance index function;
calculating the local gradient of each layer, and further updating a first parameter set of the fuzzy neural network parameters;
obtaining and updating a second parameter set of the intelligent PID controller by using the updated first parameter set;
judging whether the online fine tuning of the back propagation algorithm reaches the maximum time length TmaxIf the judgment result is negative, returning to carry out the fuzzy processing again; and if so, obtaining the optimal second parameter set after finishing the repeated online back propagation algorithm fine adjustment.
6. The cabinet temperature control method of the in-vehicle refrigerator as claimed in claim 5, wherein the back propagation algorithm selects a mean square error loss function as a performance index function:
where e (n) is the control error for each iteration step. By a symbolic function
Approximate substitution
The error is compensated by adjusting learning rate eta, and the calculated network weight omegaijMembership function base width bijCenter value of membership function cijComprises the following steps:
wherein the learning rate eta is eta1＝η2＝η3> 0, alpha is a momentum factor.
7. The method for controlling the box body temperature of the vehicle-mounted refrigerator according to claim 1, wherein the triggering of the intelligent PID controller to control the box body temperature of the current vehicle-mounted refrigerator by using the optimal second parameter set determined based on the combination of an offline algorithm and an online algorithm specifically comprises:
reading a target temperature r of a corresponding mode set for a cabinet system of an in-vehicle refrigeratorin(n)；
Detecting the current box body temperature y of the vehicle-mounted refrigeratorout(t) and let yout(n)＝TnWherein n is the number of refrigeration/heating control cycles;
comparing the current box temperature TnAnd the target temperature TsThe difference between the temperature difference and the artificially set temperature variation delta T,
in the cooling mode: when T isn≥(Ts_c+ Δ T), working according to the maximum refrigeration capacity; when | Tn-Ts_cWhen | < delta T, using the optimal second parameter set self-tuning algorithm of the improved intelligent PID controller based on the fuzzy neural network to enable TnQuickly reach Ts_cAnd remain unchanged; when T isn≤(Ts_cAt) then stop output, wait for temperature return, when Tn≥(Ts_c+ delta T), starting to work according to the maximum refrigerating capacity;
in the heating mode: when T isn≤(Ts_h-delta T), working according to the maximum heating capacity; when | Tn-Ts_hWhen | < delta T, using the optimal second parameter set self-tuning algorithm of the improved intelligent PID controller based on the fuzzy neural network to enable TnQuickly reach Ts_h and keeping unchanged; when T isn≤(Ts_h+ Δ T), stop output, wait for cooling, when Tn≤(Ts_h-delta T), starting to work according to the maximum heating capacity;
and working according to the determined actual output refrigerating/heating quantity of the semiconductor refrigerating sheet in the current control period, and returning to repeatedly execute control when the next period is up.
8. A box temperature control device of an on-vehicle refrigerator is characterized by comprising:
the off-line algorithm module is used for determining a first parameter set of the fuzzy neural network by utilizing the fuzzy neural network constructed for the intelligent PID controller and taking a target temperature value set for the vehicle-mounted refrigerator and an actual temperature value of the refrigerator body as input in an off-line mode;
the online algorithm module is used for taking the determined first parameter set as an initial value of a back propagation algorithm, carrying out fuzzy reasoning on the fuzzy neural network in an online mode, and determining an optimal second parameter set of the intelligent PID controller, wherein the optimal second parameter set comprises proportional, integral and differential control parameters for reflecting the intelligent PID controller;
and the PID control module triggers the intelligent PID controller to control the temperature of the box body of the current vehicle-mounted refrigerator by using the optimal second parameter set determined based on the combination of an off-line algorithm and an on-line algorithm.
9. An in-vehicle refrigerator, characterized by comprising: the box body, the semiconductor refrigeration piece, the inner radiating fin, the outer radiating fin, the inner circulating fan, the outer radiating fan, the temperature sensor, the control module, the partition plate and the protective cover; wherein the control module is the box temperature control device of the vehicle-mounted refrigerator of claim 8, or the control module comprises the box temperature control device of the vehicle-mounted refrigerator of claim 8;
the box body is a heat preservation box body, and each surface forming the box body is provided with a foaming heat preservation layer; arranging a partition board in the box body, dividing the inner space of the box body into an installation chamber and a storage chamber by the partition board, arranging an air outlet and an air return opening on the partition board, arranging an installation window communicated with the installation chamber on one side wall of the box body, arranging a protective cover on the outer side of the box body, and arranging a heat dissipation air opening on the protective cover;
the semiconductor refrigerating sheet is arranged on the mounting window of the box body; the inner radiating fins are arranged in the mounting chamber and are in heat transfer contact with one side surface of the semiconductor refrigerating sheet, and the outer radiating sheets are arranged on the outer side of the box body and are in heat transfer contact with the other side surface of the semiconductor refrigerating sheet; the internal circulation fan is arranged in the installation chamber, faces the air outlet and is used for blowing low-temperature/high-temperature air flow in the installation chamber into the storage chamber; the outer heat dissipation fan is arranged on the outer side of the box body, faces the outer heat dissipation sheet, is arranged in the protective cover and is opposite to the heat dissipation air inlet, and is used for providing cooling/back-heating air flow for the outer heat dissipation sheet;
the temperature sensor is arranged in the box body, is arranged in the installation chamber at a position opposite to the air return opening, is used for acquiring the temperature of the box body, and is in signal connection with the control module; the control module is in control connection with the semiconductor refrigeration sheet, the internal circulation fan and the external heat dissipation fan in a wired connection mode; the control module, the outer heat dissipation sheet and the outer heat dissipation fan are all arranged in the protective cover.
10. The in-vehicle refrigerator of claim 9, wherein the inner fin and the outer fin are fin type fins; and/or;
and heat-conducting silicone grease layers are arranged between the inner radiating fins and the semiconductor refrigerating fins and between the outer radiating fins and the semiconductor refrigerating fins.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111239207.0A CN113885609B (en) | 2021-10-25 | 2021-10-25 | Method and device for controlling temperature of refrigerator body of vehicle-mounted refrigerator and vehicle-mounted refrigerator |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111239207.0A CN113885609B (en) | 2021-10-25 | 2021-10-25 | Method and device for controlling temperature of refrigerator body of vehicle-mounted refrigerator and vehicle-mounted refrigerator |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113885609A true CN113885609A (en) | 2022-01-04 |
| CN113885609B CN113885609B (en) | 2023-04-25 |
Family
ID=79013547
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111239207.0A Active CN113885609B (en) | 2021-10-25 | 2021-10-25 | Method and device for controlling temperature of refrigerator body of vehicle-mounted refrigerator and vehicle-mounted refrigerator |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113885609B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116203342A (en) * | 2023-04-27 | 2023-06-02 | 成都中航华测科技有限公司 | Temperature control method for device testing process based on high-low temperature experiment box |
| CN117308499A (en) * | 2023-09-25 | 2023-12-29 | 广东英得尔实业发展有限公司 | Refrigerating and heating mode switching method of vehicle-mounted refrigerator |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104331097A (en) * | 2014-03-28 | 2015-02-04 | 海尔集团公司 | Semiconductor refrigeration refrigerator and temperature control method thereof |
| CN107621118A (en) * | 2017-09-29 | 2018-01-23 | 南京工业大学 | A kind of semiconductor refrigerating hutch suitable for cold chain transportation |
| CN107704985A (en) * | 2017-08-11 | 2018-02-16 | 浙江工业大学 | A kind of differential evolution Flexible Workshop Optimization Scheduling of dynamic strategy |
| CN107807530A (en) * | 2017-11-30 | 2018-03-16 | 黄力 | A kind of forcing press PID control system based on intelligent fuzzy neural network algorithm |
| CN110779268A (en) * | 2019-10-25 | 2020-02-11 | 长虹美菱股份有限公司 | Refrigerator internal temperature control system algorithm |
| CN111661456A (en) * | 2020-05-21 | 2020-09-15 | 四川虹美智能科技有限公司 | Air-cooled cosmetic storage case and control method for air-cooled cosmetic storage case |
-
2021
- 2021-10-25 CN CN202111239207.0A patent/CN113885609B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104331097A (en) * | 2014-03-28 | 2015-02-04 | 海尔集团公司 | Semiconductor refrigeration refrigerator and temperature control method thereof |
| CN107704985A (en) * | 2017-08-11 | 2018-02-16 | 浙江工业大学 | A kind of differential evolution Flexible Workshop Optimization Scheduling of dynamic strategy |
| CN107621118A (en) * | 2017-09-29 | 2018-01-23 | 南京工业大学 | A kind of semiconductor refrigerating hutch suitable for cold chain transportation |
| CN107807530A (en) * | 2017-11-30 | 2018-03-16 | 黄力 | A kind of forcing press PID control system based on intelligent fuzzy neural network algorithm |
| CN110779268A (en) * | 2019-10-25 | 2020-02-11 | 长虹美菱股份有限公司 | Refrigerator internal temperature control system algorithm |
| CN111661456A (en) * | 2020-05-21 | 2020-09-15 | 四川虹美智能科技有限公司 | Air-cooled cosmetic storage case and control method for air-cooled cosmetic storage case |
Non-Patent Citations (2)
| Title |
|---|
| 王安敏等: "医用低温冰箱温度控制系统的设计", 《计算机与数字工程》 * |
| 白国振等: "基于改进模糊神经网络的PID参数自整定", 《计算机应用研究》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116203342A (en) * | 2023-04-27 | 2023-06-02 | 成都中航华测科技有限公司 | Temperature control method for device testing process based on high-low temperature experiment box |
| CN117308499A (en) * | 2023-09-25 | 2023-12-29 | 广东英得尔实业发展有限公司 | Refrigerating and heating mode switching method of vehicle-mounted refrigerator |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113885609B (en) | 2023-04-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10845771B2 (en) | Automated method of generalized building automation from atomic physical models and control loops thereof | |
| US11391484B2 (en) | Building control system with constraint generation using artificial intelligence model | |
| CN113885609B (en) | Method and device for controlling temperature of refrigerator body of vehicle-mounted refrigerator and vehicle-mounted refrigerator | |
| Xi et al. | Support vector regression model predictive control on a HVAC plant | |
| Noye et al. | A review of advanced ground source heat pump control: Artificial intelligence for autonomous and adaptive control | |
| Swider et al. | Modelling of vapour-compression liquid chillers with neural networks | |
| Amarasinghe et al. | Artificial neural networks based thermal energy storage control for buildings | |
| Huang et al. | Multi-zone temperature prediction in a commercial building using artificial neural network model | |
| CN115774929A (en) | Electric automobile thermal management control system and method considering energy conservation and battery life | |
| Kusiak et al. | Reheat optimization of the variable-air-volume box | |
| JP2016051213A (en) | Electronic apparatus temperature management system, temperature management method, and temperature management program | |
| Wan et al. | Intelligent rack-level cooling management in data centers with active ventilation tiles: A deep reinforcement learning approach | |
| Wang et al. | A hyperparameter optimization algorithm for the LSTM temperature prediction model in data center | |
| Yousefi et al. | A hybrid neuro-fuzzy approach for greenhouse climate modeling | |
| CN112820976B (en) | Battery heat exchange fan system of electric vehicle and control method thereof | |
| CN116928944B (en) | Multi-partition refrigeration house operation control method and system | |
| de Oliveira et al. | Dynamic online optimization of a house heating system in a fluctuating energy price scenario | |
| JP2015162098A (en) | air-conditioning control system and air-conditioning control method | |
| CN116954329A (en) | Method, device, equipment, medium and program product for regulating state of refrigeration system | |
| Avila-Miranda et al. | An optimal and intelligent control strategy to ventilate a greenhouse | |
| Hussain et al. | A multi‐objective evolutionary optimization of fuzzy controller for energy conservation in air conditioning systems | |
| Singh et al. | Multi parameter estimation in an induced draft cooling tower using genetic algorithm | |
| Hreinsson et al. | On aggregating thermostatically controlled loads based on energy losses | |
| CN113419425A (en) | Intelligent regulation and control method for high-speed electric spindle water cooling system | |
| CN111637614A (en) | Intelligent control method for data center active ventilation floor |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |