JPS6240721B2 - - Google Patents

Description

[Detailed description of the invention]

The present invention relates to a control computing unit including a microprocessor used for process control. In the field of process control, in addition to the four arithmetic operations, there are functions unique to control operations such as selectors and limiters, and conventionally, analog arithmetic units with a single function have been combined to satisfy the overall function. In recent years, control systems have become more complex due to issues such as resource conservation and energy conservation, and constants, coefficients, or arithmetic expressions are often adjusted to match the actual process. For this reason, analog calculations have problems such as decreased accuracy, limited functionality, complicated wiring, and decreased reliability. To deal with this, a computer-based control method (DDC) was put into practical use, but due to its reliability and cost, its use was limited to only a few, and it was only with recent microprocessor application technology that it was applied to small-scale instrumentation. Now I can do it. In signal processing devices using microprocessors,
Semiconductor memories are used as memory elements for storing programs because they are easy to use in terms of drive circuits and miniaturization. RAM (Random Access Memory), which can be read and written freely, is used to allow programmers to freely assemble programs. RAM is volatile;
Since the contents are lost when the power is turned off, countermeasures such as battery backup have been taken, but noise-resistant RAM measures and battery maintenance are also required, as well as a master file that does not erase programs. LOAD／
A SAVE device is essential. In this way, in conventional microprocessor-based signal processing devices, the safety of the programs assembled in the RAM inside the device,
Service tools are required to improve reliability, and the price of the system increases in exchange for the convenience of programmability. On the other hand, software production is essential for computer-based control systems. Since the user does not want to disclose control know-how, he would like to produce the software, but since it requires a huge amount of man-hours, he has no choice but to order it from the manufacturer. For manufacturers, there are problems such as an increased burden on software and an increase in average prices. For this reason, software languages such as FORTRAN, BASIC problem language (POL), and FILL IN THE FORM format have been developed, but all of them require a considerable amount of grammar and hardware to create programs. , it is not in a state where it can be used freely by all users, and it cannot be said that it is an optimal language for low-cost equipment such as microprocessor application systems. The present invention divides a program into a fixed system program and an application program that changes for each job, stores them in separate ROMs (read-only memories), and executes control operations in the ROM for the system program. Various functions such as arithmetic operations, logical operations, and dynamic operations necessary for the calculation are stored as a calculation library, and the procedure for executing which functions in this calculation library and in what order is specified in the application program. In this way, a novel control arithmetic unit without the above-mentioned drawbacks has been realized. FIG. 1 is a block diagram showing one embodiment of the arithmetic unit of the present invention, and FIG. 2 is a front configuration diagram thereof. In the figure, reference numeral 1 denotes a reference voltage generating section, which provides a reference value Er of an analog signal (for example, DC5V). 2 is an input buffer amplifier, which receives DC1 from the outside.
The input values Ei ₁ to _{Ei 4} of analog signals such as ˜5V are impedance-converted, respectively. 3 is a variable calculation constant setting circuit, which has 8 potentiometers.
It has PM ₁ to PM ₈ and provides set values Ep ₁ to _{Ep 8} of various variable calculation constants necessary for calculation processing. And each potentiometer PM ₁ ~
As shown in Figure 2, the PM ₈ volume is arranged so that it can be set using a screwdriver from the front of the calculator. A storage compartment for the driver DV for setting operations is provided at the lower right of the front of the calculator. 4 is an input multiplexer, which has a reference value Er and input values Ei ₁ to _{Ei 4.}
and a switch for sequentially selecting set values Ep ₁ to _{Ep 8} . 5 is a signal conversion circuit,
It performs A/D conversion and D/A conversion.
Reference numeral 6 denotes an output multiplexer, which is composed of switches for sequentially selecting output values of analog signals obtained by D/A conversion. Reference numeral 7 denotes an output hold circuit, which is a capacitor hold type amplifier, and holds the output signals Eo ₁ to Eo ₄ so that they do not vary during each scan period. A microprocessor 8 (hereinafter referred to as CPU) controls the input/output multiplexer, A/D conversion and D/A conversion of the signal conversion circuit, performs digital calculations, self-diagnosis, etc. 9 is RAM, which is a memory used for temporary storage of data, etc., and includes an input register,
It has several dedicated registers such as output registers and temporary storage registers. 10 is for system
ROM is a memory that stores system programs. The system block consists of a management program for managing the operation of the entire computing unit, an input/output signal processing program, an arithmetic processing program, a data display program, a self-diagnosis program, and operations (arithmetic operations, logical operations, dynamic operations, etc.)
It is a fixed program composed of libraries. 11 is the application ROM,
This is memory in which application programs are stored. The application program is a program created for each job, including information regarding the calculation procedure, fixed calculation constants necessary therefor, and display register. Therefore, the ROM 11 is mounted on the board with a socket or the like so that it can be replaced. 1
2 is an I/O interface that exchanges digital signals between the CPU and peripheral circuits. 13 is a timer that manages the scanning period (100ms) of calculations, and a clock using a crystal oscillator is used as the basic clock.
Sends a timer interrupt signal to the CPU every 100ms.
14 is a watch dog timer, which operates when the CPU does not operate normally due to a CPU abnormality or program abnormality.If an abnormality is detected, it outputs a fail signal to the outside and also Display is performed using LAMP ₁ installed in the front. Reference numeral 15 denotes a KEY/DISPLAY interface, which reads the input of a selection switch, which will be described later, and drives the display section. Reference numeral 16 denotes a power supply unit that converts power supplied from the outside into a voltage required by the internal circuit. 17 is a voltage level detection circuit that sends a reset signal to the CPU when the power is turned on to ensure initialization of peripheral circuits.
Give RS. Reference numeral 18 denotes a display data selection switch, which is a push button switch for selecting display data in an UP/DOWN manner, and is provided in front of the computing unit as shown in FIG. The selection operation is performed while monitoring the channel selection display lamps LED ₁ to _{LED 16} provided in front of the computing unit. 1
9 is the data display, a 5-digit 7-segment LED
It is installed in front of the computing unit and displays the input value, output value, and set value as an industrial quantity using a code plus a 4-digit number. Information regarding the range, unit, etc. of this industrial quantity is written on the data label DL provided on the front of the calculator. 20 is a DI interface which converts an external digital input signal (contact signal) Di into a logic level. 21 is a DO interface which converts the digital output signal Do into a relay contact or a transistor contact and outputs it. In the arithmetic unit of the present invention configured in this way,
All internal operations are performed under the control of the CPU, and the system program that specifies the control contents is
It is included in ROM10. First, when the power is turned on, the peripheral circuits are initialized for a certain period of time. Next, the reference value Er, the input values Ei ₁ to Ei ₄ and the set values Ep ₁ to _{Ep 8} of the variable calculation constants are sequentially selected by the input multiplexer 4, and A/D conversion is performed by the signal conversion circuit 5. Each A/D converted signal is normalized and stored in its own dedicated input register. On the other hand, digital inputs (contact signals) are also read and processed into dedicated input registers. Once all the input information has been stored in the input register, the calculations are performed by combining the calculation libraries such as arithmetic operations, logical operations, and dynamic operations in the system program according to the calculation procedure given by the application program. Processing is executed. At this time, as shown in FIG. 3, the input registers in the RAM (X ₁ to _{X 4} for analog input, P ₁ to _{P 8} for variable constants, and Di for digital input) and the fixed constant registers K ₁ to K in ROM 11 In addition to ₁₆ , calculation registers S ₁ to 1 in the CPU
_S3 and temporary storage registers _T1 - _T4 in RAM are used. The operation result (output signal) is entered into the output register in RAM. As shown in FIG. 3, the output registers include analog output registers _Y1 to _Y4 and digital output register _D0 .
The contents of the analog output registers are read out in sequence and subjected to D/A conversion by the signal conversion circuit 5. Each D/
The A-converted result is switched by the output multiplexer 6 and applied to the output hold amplifier 7, where it is held. The contents of the digital output register are also read out and output as a contact signal via the DO interface 21. The steps from reading the input signal to calculation and output are as follows.
Since it is executed repeatedly at a cycle of 0.1 seconds, the response is almost the same as that of other analog computing units. During this time, the input signals, variable constant settings, and output signal values stored in the input register and output register are all selectively displayed numerically on the data display 19 on the front. The display format is not only 0 to 100% values, but also industrial quantities such as 0 to 2000 Nm 3 /h, such as 0 to 2000 Nm ³ /h, can be specified by the application program. Next, the calculation steps given in the application program are instructions for reading inputs, constants, etc.
It consists of three types: the LOAD instruction), the instruction to execute an operation (hereinafter referred to as FUNC instruction), and the instruction to output the result of operation (hereinafter referred to as STORE instruction), as well as an instruction to branch the program as an auxiliary instruction (hereinafter referred to as STORE instruction). There are two commands: the GOTO command) and the command that indicates the end of the program (hereinafter referred to as the END command). The operation registers S ₁ to _{S 3} used for all operations by these main instructions function as follows.

【table】

[Table] That is, the LOAD instruction stores data such as inputs and constants in _S1 , and at the same time has a pushdown function that shifts the previously stored data to the next register. The FUNC instruction has a pop-up function that executes an operation using the data input to _S1 to _S3 , stores the result in _S1 , and carries forward the data not used in the operation. In the STORE instruction, S ₁ which is the operation result
The data is output to the register or temporary storage register, but the contents of the calculation registers S ₁ to _{S 3} remain unchanged. All analog input/output, variable constants, etc. are stored as normalized data in dedicated registers, so all signals and information, including fixed constants and temporary memory data,
You can read and read at STORE. Also, for the FUNC instruction,
As shown in FIG. 4, in addition to the four arithmetic operations, there are signal conversion functions, selector functions, limiter functions, dynamic functions, condition determination functions, and digital output functions necessary for control calculations. Therefore, in the control system represented by the instrumentation flowchart of FIG. 5, the calculation procedure is as shown in FIG. Since the calculation results are always popped up in _S1 , the calculation can be continued immediately by simply loading the additional data needed for the next calculation. Most of the programs are of two types: LOAD and FUNC instructions, making them extremely simple. In this way, the arithmetic procedure allows complex operations to be executed simply by combining the three types of instructions: LOAD, FUNC, and STORE. Moreover, the three types of instructions, LOAD, FUNC, and STORE, can be performed simply by combining the three types of instructions, LOAD, FUNC, and STORE. It is easy to understand because it is a flow of instructions that follow the operating procedure, and it is easy to convert instrumentation flowcharts, which are often written as signal flows, into programs (sometimes difficult to express using formulas such as FORTRAN or BASIC). . Therefore, program creators do not need to have any knowledge of computer software, and can calculate process inputs and outputs using procedures comparable to those of an analog calculator while viewing the instrumentation procedure. It only requires a small-scale conversion into an instruction code, and tools using large computers such as general compilers and generators are not required. In particular, as shown in the keyboard panel diagram of FIG. 7, functions specific to control calculations can be realized by one-touch key operations, making it possible to provide the user with a programming tool.
Therefore, the user can easily create a program without disclosing control know-how. Note that the condition judgment function is used for digital input/output.
It branches the program based on ON/OFF and analog signal level comparisons, and can be used in combination with the GOTO instruction to execute sequence operations. In addition, in this embodiment, the reference value Er is
What is converted and read is that the signal is normalized by removing the influence caused by variations in the characteristics of the hardware of the signal conversion circuit 5 by performing digital calculations using the conversion result. It's for a reason. Furthermore, LAMP ₂ installed in front of the computing unit is an abnormality display lamp that lights up when the input signals Ei ₁ to _{Ei 4} exceed a level of -5% to 105% or when the signal conversion circuit 5 becomes abnormal. It is. As explained above, in the present invention, programs are divided into a fixed system program and an application program that changes for each job, and each program is stored in a separate ROM.
Various functions such as arithmetic operations, logical operations, and dynamic operations necessary to execute control operations are stored in the ROM as a calculation library, and it is possible to determine which functions in this calculation library are to be executed in what order. Since the procedure is specified in the application program, all you have to do is replace the ROM for the application program.
A control computing unit having a desired control purpose can be easily realized. Moreover, the application program only needs to specify the procedure for executing which functions in the calculation library and in what order, and a control calculation unit that can perform procedures comparable to analog calculations can be obtained.

[Brief explanation of the drawing]

Fig. 1 is a block diagram showing one embodiment of the arithmetic unit of the present invention, Fig. 2 is a front configuration diagram thereof, Fig. 3 is a configuration diagram of a register in the arithmetic unit of the present invention, and Fig. 4 is an operation diagram of the inventive arithmetic unit. A list of functions, FIG. 5 is an instrumentation flowchart showing an example of a control system to which the present invention is applied, FIG. 6 is an example of its program, and FIG. 7 is an example of the keyboard of the programming tool used in the present invention. It is a panel diagram. 5... Signal converter, 8... Microprocessor, 9... RAM, 10... ROM for system program, 12... ROM for application program, _S1 , _S2 , _S3 ... Arithmetic register.

Claims (1)

[Claims] 1. A system program ROM in which a fixed system program is stored as a memory connected to a microprocessor via a data bus.
and an application program ROM that is exchanged for each job and stores an application program corresponding to that job, and the system program ROM has arithmetic operations and logic necessary for executing control operations. Various functions such as calculations and dynamic calculations are stored as a calculation library, and the procedure for executing which functions in this calculation library and in what order is stored in the application program ROM. A control computing unit characterized by being configured to execute control calculations determined by a program. 2. The application program stored in the application program ROM is composed of a combination of a read instruction with a push-down function, and a plurality of arithmetic instructions with an output instruction and a pop-up function. 2. The control arithmetic unit according to claim 1, wherein the control arithmetic unit is provided corresponding to various functions such as arithmetic operations, logical operations, and dynamic operations of the arithmetic library.