RU2422873C2 - Method and apparatus for rapid dynamic analysis of states of multi-parameter object - Google Patents

Abstract

FIELD: information technology.

SUBSTANCE: disclosed is a method for rapid dynamic analysis of states of a multi-parameter object, involving presentation of results of complete analysis of processes for dynamically optimum control of the object in form of a cognitive icon of optimum control equivalents, which is formed in the address space of the code matrix of read-only memory. Said icon is then used for real-time synthesis of energy-saving control effects for any changes in functioning states of the multi-parameter object in the estimated range with given accuracy of the synthesised effects which are regulated by the reference signal error. Disclosed also is a device for rapid dynamic analysis of states of a multi-parameter object.

EFFECT: higher metrological and technological efficiency owing to automatic optimum control of energy saving.

2 cl, 4 dwg, 3 tbl

Description

The group of inventions relates to the field of structural pattern recognition and can be used in automated systems for the operational diagnostics of the technical and functional states of a multi-parameter object according to measurement information.

A known method and device for operational dynamic analysis of the states of a multi-parameter object of a decentralized type [see Prince Electrical measurements. / Ed. V.N. Malinovsky. - M .: Energoizdat, 1982. - S.377-378]. In the method, all signals are differentiated by individual functions that perform a predetermined operation on the information signal. The device is organized according to the main structure connecting the object with a computer and a translator through a communication device.

The disadvantages of these solutions are the dispersal of information across different individual functional structures with a wide range of products, the excessive complexity of the device, the inability to quickly analyze the resulting variables dynamically due to the independence of the information channels from each other.

Also known is a method and device for centralized dynamic control and analysis of the states of a multi-parameter object [see Prince Theory of automatic control./ Ed. Yu.M. Solomentseva. - M .: Higher school, 2000. - S.202-204]. The method consists in program-controlled monitoring and analysis of the state of the object based on the library of programs, together with the dispatcher program in a predetermined order or depending on the current values of the technological parameters, with the subsequent selection of one of the work programs available in the library. The device is organized on the basis of an information-measuring system with a ring structure of sequentially connected data input interface, a computer, an executive converter, measuring transducers, and display means operating in accordance with the commands of an operator or a dispatcher program.

Their disadvantages are the low efficiency in executing the commands and procedures necessary to evaluate the system and its correction, expensive software and hardware, and the complexity of the operational integrated assessment of incoming information.

The prototype adopted the method of operational dynamic analysis of the states of a multi-parameter object [see Patent No. 2134897 (RF), PMK G05B 19/408, G06F 17/40, published 1999.08.20], which consists in measuring and evaluating the integral state of a multi-parameter object from the graphic image of a cognitive matrix, converting the results of the tolerance assessment of heterogeneous dynamic parameters into corresponding information signals with a generalization over the entire set of parameters in a given time interval and determining the relative magnitude and nature of the change in the integral state of the multiparameter object.

For a prototype of the implementation of the method adopted device dynamic dynamic analysis of the states of a multi-parameter object [see Prince Theory of automatic control. / Ed. Yu.M. Solomentseva. - M .: Higher school, 2000. - P.207], containing a control device, a pulse generator connected to the clock input of the pulse counter, an amplifier, a power supply, a multiparameter object, a digital-to-analog converter, the output of which is connected through an amplifier to the actuators of the multiparameter object the control output of the tolerance control of which is connected to the control inputs of the power consumption modes of the control unit, the control output of the power consumption of which is connected to the corresponding input of the pit block Niya. The operator, using a computer, controls the output parameters of the process and varies the ranges of permissible changes in the effects on a multi-parameter object.

The disadvantages of the prototypes are low metrological and technological efficiency, as well as the lack of energy efficiency of technological equipment with a minimum of energy costs due to the impossibility of automatic decision-making in real time, including the implementation of adaptive control actions based on the analysis of the color code matrix of the state of the investigated object with a minimum of costs energy.

The technical task of the proposed method and device is to improve metrological and technological efficiency due to automatic optimal energy-saving control, which uses energy-saving control actions for real-time synthesis for any changes in the state of a multi-parameter object in the analyzed range with a given accuracy of the synthesized actions, regulated by the error of the model signals.

The technical task is achieved in that:

P1. In the method of operational dynamic analysis of the states of a multiparameter object, which consists in measuring and evaluating the general condition of a multiparameter object using a graphical image of a cognitive matrix {Ф} = {L ₁ * L ₂ }, in contrast to the prototype, they present the results of a complete analysis of the dynamics of the optimal control of the object described a model of a double aperiodic link with a minimized functional for energy costs, in the form of a cognitive graphic image of the equivalents of optimal control, which forms comfort in the address space of the code matrix permanent memory, and then used for the synthesis in real-time power saving control actions at any changes states functioning multiparameter object in the estimated range with a specified precision synthesized influences regulated error model signals, where L ₁ - a plurality of control actions , L ₂ - measures to assess management costs.

P2. The device according to claim 1, containing a control unit, a pulse generator connected to the clock input of the pulse counter, an amplifier, a power supply, a multi-parameter object, a digital-to-analog converter, the output of which is connected through an amplifier to the actuators of the object, the control output of which is controlled by tolerance control, and is connected to the control inputs power consumption modes of the control device, the control energy output of which is connected to the corresponding input of the power supply, in contrast to the prototype it contains a permanent storage device and an encoder connected at the inputs to the output of the sensors of the object, and at the outputs to the higher address bits of the permanent memory connected by the lower address bits to the outputs of the counter, and the output bus of the permanent storage device is connected to the information inputs of the digital-to-analog converter the input of which is connected to the control information output of the control device, the control clock output of which is associated with yuschim input of the pulse generator.

The method and device are presented in figures 1-4.

The essence of the proposed method of operational dynamic analysis of the states of a multi-parameter object (see Figs. 1-4) consists in measuring and evaluating the general state of the object, information about which is presented in the form of a cognitive graphic image of the optimal control equivalents {Ф} = {L ₁ * L ₂ } (Fig.1), formed according to the diagnostic information (Fig.2, curve z) with the subsequent synthesis of energy-saving control actions in real time (Fig.3, curve u *).

The results of a complete analysis of the processes of the dynamics of optimal control of an object are presented in the form of equivalents (Fig. 1), which form a constant storage device (ROM) in the address space of the code matrix. This image is found a priori from a complete analysis of the processes of the dynamics of optimal control of an object described by a model of a double aperiodic link with a minimized functional in terms of energy consumption. The use of a cognitive graphic image of the optimal control equivalents makes it possible to exclude the optimal control analysis procedure in the dynamic analysis of the state of a multi-parameter object in real time, thereby significantly lowering the requirements for technical means that implement control.

We explain the proposed method by the example of optimal control of the dynamic process of electrical equipment. The dynamics of electric heaters and electric motors can be described by a system of differential equations

where z ₁ and z ₂ are phase coordinates, parameters a ₁ and a ₂ characterize the inertia of the control object, parameter b is the gain of the control action u (t).

в конечное

, при заданной скорости изменения данной фазовой координаты от

до

, необходима сложная двухэтапная процедура, состоящая из анализа и синтеза оптимального управления в реальном масштабе времени.To determine the optimal control actions u * that implement the optimal energy-saving control for the transfer of a multi-parameter object from the initial state

in the final

, at a given rate of change of a given phase coordinate from

before

, a complex two-stage procedure is required, consisting of the analysis and synthesis of optimal control in real time.

The analysis of optimal control covers a wide range of problems associated with research into the existence of a solution, stability, determining the possible types of functions of optimal control, and many other problems. Performing the optimal control analysis procedure requires complex algorithmic and mathematical calculations and, as a result, expensive high-performance microprocessor-based hardware. This is an obstacle to the operational dynamic analysis of the states of a multiparameter object.

The results of the analysis of optimal control on a multitude of functioning states serve as the basis for constructing a cognitive graphic image, which will be used in the future for constructing energy-saving control devices for various dynamic objects. To perform the analysis of optimal control in the system, the initial data for solving the problem are set, presented in the form of an array of details

The result of the analysis is to obtain a cognitive image in the form of a color code matrix {Ф} = {L ₁ * L ₂ } of the areas of existence of controls (Fig. 1). The type of minimized functionality is preserved over the entire control time interval

the control action at a given time interval is limited

where u _n is the lower threshold of the control action; u _in - the upper threshold of the control action.

Therefore, the control is limited, at each moment of time, the ends of the trajectory of the change in phase coordinates are fixed and the time interval is fixed.

From the obtained on the basis of a full analysis of the cognitive graphic image (figure 1), the areas for which there is optimal control will be areas I, II, III, IV, V, VI, VII.

The synthesis of control actions is carried out according to the results of a complete analysis a priori and presented in the form of a cognitive graphic image {Ф} = {L ₁ * L ₂ }, formed in the address space of the code matrix L ₁ * L ₂ ROM of the control device, according to the array of initial details optimal control problems (2).

As a result of the optimal control synthesis procedure, we determine the region (Fig. 1) into which point S with coordinates (L ₁ , L ₂ ) falls. Each area corresponds to a certain type of control function.

If the optimal control is not found, i.e. If the coordinates of the point S = (L ₁ , L ₂ ) do not belong to the regions of the I-VII cognitive graphic image, then the automatic control device is transferred from the optimal to the traditional mode of operation for a multi-parameter object.

Figure 4 presents the functional diagram of the device, which includes a multi-parameter object 1, control unit 2, pulse generator 3, counter 4, ROM 5, digital-to-analog converter (DAC) 6, amplifier 7, encoder 8, power supply 9.

DAC 6 is connected through an amplifier 7 to the actuators of the multiparameter object 1. The control output of the tolerance control of the multiparameter object 1 is connected to the control inputs of the power consumption modes of the control unit 2, its control output of power consumption is connected to the corresponding input of the power supply 9. The encoder 8 is connected to the inputs of the sensors multiparameter object 1, and the outputs - with the highest address bits of the ROM 5, connected by the lower address bits with the outputs of the counter 4. Output I tire ROM 5 is connected to data inputs of the DAC 6, a control input coupled to the control data output of the control unit 2, its clock output connected to a corresponding input of the pulse generator 3.

The inventive method is implemented by a device for operational dynamic analysis of the states of a multi-parameter object as follows.

From the control output of the tolerance control of a multi-parameter object 1 to the control input of the power consumption modes of the control unit 2, power signals of the device circuit are transferred, which convert it to the on, off or standby modes, and from the control output of the power consumption of the control unit 2 to the input of the unit power 9 signals are given corresponding to the set operating modes of the device - "on", "off" or "standby". The power supply 9 provides power to all functional blocks 1-8 of the device in accordance with a given mode. “On” - power is supplied to all units of the device, “off” - the device is de-energized, “standby” - the device is switched to low power mode, in which only the control unit 2 operates. In this case, the device is able to enter the operating mode much faster, than from the "off" state. From the clock output of the control unit 2 to the corresponding input of the pulse generator 3 receives the signal on or off of the pulse generator 3, the clock pulses of which are fed to the clock input of the counter 4.

From the outputs of the sensors of the multiparameter object 1, the signals corresponding to the array of details (2) of the optimal control problem are sent to the inputs of the encoder 8. From the outputs of the encoder 8, the code sequence corresponding to the array of details is fed to the senior address bits A _{n of the} ROM 5, which are the address group of one of the seven zones I-VII of the cognitive graphic image (Fig. 1) of the optimal control areas. Each zone corresponds to a certain type of optimal control function.

The formation of the optimal control signal for a multi-parameter object 1 is carried out by enumerating 4 addresses of the lowest address group A _m ROM 5, where the optimal control function corresponding to the selected area of the cognitive graphic image (Fig. 1) of the address space A _{n is} recorded. On the output bus of the ROM 5, N (A _m + A _n ) code combinations are generated that correspond to the control action on the multi-parameter object 1, which are converted into an analog PAP 6 signal and normalized by the amplifier 7.

If the optimal control is not found, i.e. the coordinates of the point L = (L ₁ , L ₂ ) do not belong to any of the seven areas of the cognitive graphic image (Fig. 1), a signal is sent from the control information output of the control unit 2 to the control input of the DAC 6, which switches the device into the traditional control mode.

в конечное

с минимумом затрат энергии.Let us prove the effectiveness of the proposed method and device for operational dynamic analysis of the states of a multi-parameter object relative to the prototype. Theoretical studies and practical results show that with optimal control, the reduction in energy costs can reach from 5% to 40% compared to traditionally used control actions [see Adzhiev M.E. Energy Saving Technologies. - M .: Energoatomizdat, 1990.64 p .; Arakelov V.E., Kremer A.I. Methodological issues of energy saving. - M .: Energoatomizdat, 1990. 188 pp.] As an example, consider the transfer of a thermal object - a heating plate from its initial state

in the final

with a minimum of energy costs.

An array of details for the analysis of energy-saving management are given in table 1.

Table number 1 Initial data of the experiment Parameter Value a ₁ -0.10 a ₂ -1.10 b 1.00 t ₀ 0 t _k 5.00 u _n 0 u _in 200.00

15.00

320.00

0

140 I _e 104000

I _E is the functional (limit) of energy expenditures under traditional control (Fig. 3, curve u). Therefore, the task is to determine such a control that will ensure the transfer of the object from the initial state to the final one with a minimum of energy consumption (2).

As a result of a complete analysis of the optimal control, we obtain a color code matrix in the form of images of the equivalents of the set of optimal control states (Fig. 1).

Next, a cognitive graphic image is formed in the address space of the code matrix of a read-only memory in the form of equivalent images of a plurality of optimal control states.

At the next stage, the synthesis operation is performed, as a result of which we obtain the coordinates of the point S (Fig. 1, table No. 2), and in accordance with the coordinates of the obtained point, the type of the control function is selected and implemented (Fig. 2, curve z * table No. 3).

Table number 2 Experiment Results Parameter Value L ₁ 0.38117 L ₂ 43,22061 Zone number IV I 65672

Figure 2 and table No. 3 shows the dependencies that display the temperature change of the heating system with traditional and optimal control.

Table number 3 Experiment Results t min Traditional management Energy saving management z, ° C u, B z ^* , ° С u ^* , B 0 15.0 200.00 15.0 51,4 0.2 18.6 200.00 15.9 51.1 0.4 28.7 200.00 18,4 51.0 0.6 43.9 200.00 22.2 51.1 0.8 63.1 200.00 27.0 51,4 one 85,4 200.00 32.6 52.0 1,2 110,2 200.00 38.8 52.9 1.4 136.9 200.00 45.6 54,4 1,6 165.0 200.00 52.9 56.3 1.8 194.2 200.00 60.5 59.0 2 224.0 200.00 68.5 62,4 2.2 254.4 200.00 77.0 66.9 2,4 285.0 200.00 85.9 72.7 2.6 320,0 - 95.4 80.0 2,8 - - 105.5 89.1 3 - - 116.5 100.6 3.2 - - 128.5 114.9 3.4 - - 141.8 132.7 3.6 - - 156.8 154.6 3.8 - - 173.8 181.7 four - - 193.4 200.00 4.2 - - 215.6 200.00 4.4 - - 239.8 200.00 4.6 - - 265.6 200.00 4.8 - - 292.4 200.00 5 - - 320,0 -

Thus, using the property of inertia of the object and a slightly longer heating time with energy-saving control, we obtain the energy consumption functional I = 65672, which, when compared with the energy consumption functional of standard control, corresponds to an energy saving of 36.85%.

We also calculate the manufacturability of the claimed device in comparison with the prototype.

where, τ ₀ is the normalized operation time (τ ₀ = 1), τ ₁ is the execution time of all operations by the claimed device, τ ₂ is the execution time of all operations by the prototype, n is the number of operations performed by the claimed device, m is the number of operations performed by the prototype , k - efficiency coefficient, η - manufacturability.

The results of a complete analysis of the processes of the dynamics of optimal control of an object, a priori found and formed in the address space of the ROM device in the form of equivalent images (Fig. 1), allow us to exclude the procedure for analyzing optimal control in the dynamic analysis of the state of a multi-parameter object in real time, thereby significantly reducing the volume required calculations and the number of functional blocks of the device in relation to the prototype. Thus, the manufacturability η of the proposed method and device will be 4 to 100 times higher than the known solutions, depending on the type of multi-parameter device and the features of its operation.

The proposed method and apparatus for the on-line dynamic analysis of the states of a multi-parameter object allows to simplify the procedure for measuring and evaluating the general state of a multi-parameter object due to the fact that a complex procedure for the analysis of optimal control is performed a priori, as a result of which the cognitive graphic image of the optimal control equivalents is written in the address space of the ROM code matrix and its definition is not required in the process of dynamic analysis and control, which ultimately increases It has energy and metrological efficiency of 36.85% and technological efficiency of 4 ÷ 100 times.

Thus, the proposed method, by presenting the results of a complete analysis of the processes of optimal control dynamics of a multi-parameter object, described by a model of a double aperiodic link with a minimized functional for energy costs, in the form of a cognitive graphic image of the optimal control equivalents generated in the address space of the code matrix of a permanent storage device used for real-time synthesis of energy-saving control actions For any changes in the functioning states of a multiparameter object in the analyzed range with a given accuracy of the synthesized influences regulated by the error of the model signals, in contrast to the known solutions, it increases the metrological and technological efficiency of multiparameter objects with a minimized functional for energy consumption. This allows you to reduce the hardware costs of the control system, due to reduced requirements for system speed, microcircuit configuration, as well as to make automatic decisions in real time, including the implementation of adaptive control actions based on the analysis of the color code state matrix of the investigated multi-parameter object with a minimum of costs energy.

Claims (2)

1. The method of operational dynamic analysis of the states of a multiparameter object, which consists in measuring and assessing the general condition of a multiparameter object using the graphic image of a cognitive matrix {Φ} = {L ₁ * L ₂ }, characterized in that they present the results of a complete analysis of the dynamics of optimal control of an object, described by the model of a double aperiodic link with a minimized functional for energy costs, in the form of a cognitive graphic image of the equivalents of optimal control, which forms into the address space of the code matrix permanent memory, and then used for the synthesis in real-time power saving control actions at any changes states functioning multiparameter object in the estimated range with a specified precision synthesized influences regulated error model signals, where L ₁ - a plurality of control actions , L ₂ - measures to assess management costs. 2. Device operational dynamic analysis of the states of a multi-parameter object, containing a control device, a pulse generator connected to the clock input of the pulse counter, an amplifier, a power supply, a multi-parameter object, a digital-to-analog converter, the output of which is connected through an amplifier to the actuators of the object, the control output of the tolerance control of which connected to the control inputs of the power consumption modes of the control device, the control output of the power consumption of which о is connected to the corresponding input of the power supply unit for switching it to power saving mode, characterized in that it further comprises a read-only memory and an encoder connected at the inputs to the output of the object’s sensors, and at the outputs to the senior address bits of the permanent memory device connected by the lower address bits with the outputs of the counter, and the output bus of the permanent storage device is connected to the information inputs of the digital-to-analog converter, the control input of which is It is connected to the control information output of the control device for switching the device for the on-line dynamic analysis of the states of a multi-parameter object, and the control clock output of the control unit is connected to the corresponding input of the pulse generator for switching it on / off.

Images