Edge computing system and method with analog quantity output control

Abstract

The invention relates to an edge computing system and method with analog quantity output control. The edge computing system comprises a data acquisition module, an edge computing module and a control output module, the data acquisition module preprocesses the acquired sensor signal to obtain sampling data; the edge calculation module executes a control algorithm based on the sampling data to generate a target control quantity; the target control quantity is in a digital signal form; the control output module can convert the received target control quantity into a control signal and output the control signal; the control signal is in an analog signal form; and the control output module and the edge calculation module are integrated on the same hardware platform. According to the edge computing system and method, the analog quantity output control response speed can be increased, and the compatibility is improved.

Description

Technical Field

[0001] This invention relates to the fields of industrial automation control and edge computing technology, and in particular to an edge computing system and method with analog output control. Background Technology

[0002] Industrial control is central to modern manufacturing and energy systems, impacting production stability and efficiency. With the rapid development of the information society, the demand for remote control continues to rise. Traditional remote industrial control typically employs a PLC + host computer architecture, but this architecture suffers from high response latency and strong reliance on the cloud. Therefore, the practice of using edge computing devices to offload some control tasks has emerged. Edge computing, as a technology that brings computing and data storage closer to the data source, has wide applications in numerous fields such as industrial automation, smart buildings, and the Internet of Things.

[0003] However, the applicant noted that existing edge computing devices primarily focus on digital I / O, often requiring external DAC modules to handle analog output demands. This increases system complexity and incurs additional hardware costs and control logic burdens. Furthermore, PLC control strategies rely on cloud-based distribution, losing decision-making and control capabilities when the network is down; their general PID control algorithms also exhibit poor adaptability under nonlinear loads. Since most devices use fixed functional modules, their scalability and flexibility are insufficient, making it impossible to flexibly adjust the number of analog channels or protocol types according to actual needs. For example, when supporting protocols such as Modbus, CAN, and EtherCAT is required, it is often necessary to replace the equipment and related hardware modules. In addition, achieving collaborative control of multiple devices is difficult, thus hindering optimal system-level operation. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides an edge computing system and method with analog output control, which can improve the response speed and compatibility of analog output control.

[0005] According to an embodiment of the present invention, in one aspect, an edge computing system with analog output control is provided, characterized in that the edge computing system comprises: A data acquisition module, wherein the data acquisition module receives one or more sensor signals through an industrial protocol interface and preprocesses the acquired sensor signals to obtain sampled data; An edge computing module executes a control algorithm based on the sampled data to generate a target control quantity; the target control quantity is in the form of a digital signal. A control output module includes a communication driver circuit, a microcontroller connected to the communication driver circuit, a digital-to-analog converter circuit connected to the microcontroller, and an output circuit connected to the digital-to-analog converter circuit. The microcontroller receives digital signals from the edge computing module via the communication driver circuit. Under the control of the microcontroller, the digital-to-analog converter circuit converts the received digital signals into analog signals, which are then output via the output circuit. The control output module can convert the received target control quantity into a control signal and output it. The control signal is in analog signal form. The data acquisition module, the edge computing module, and the control output module are connected and communicate with each other through a data interface, and the control output module and the edge computing module are integrated on the same hardware platform.

[0006] Furthermore, the digital-to-analog converter circuit includes a multi-channel DAC chip, and the output circuit includes an output terminal and a multi-channel operational amplifier filter circuit connected to the output terminal. The operational amplifier filter circuit is connected to the multi-channel DAC chip, and the number of operational amplifier filter circuits is less than or equal to the number of output channels of the multi-channel DAC chip. The microcontroller can control the digital-to-analog converter circuit to output multiple analog signals.

[0007] Furthermore, any one of the operational amplifier filter circuits includes an input resistor, a feedback resistor, and an isolation resistor, and the operational amplifier filter circuit also includes: An operational amplifier is provided, with its non-inverting input terminal connected to the output terminal of the corresponding channel of the multi-channel DAC chip via the input resistor; its inverting input terminal is grounded via a first resistor; and its output terminal is connected to the inverting input terminal via the feedback resistor. The output terminal is connected to the output terminal of the operational amplifier via the isolation resistor. The resistance of the feedback resistor is 1.5 to 3.5 times the resistance of the input resistor.

[0008] Furthermore, the control output module also includes a power supply circuit for supplying power to each circuit in the control output module, the power supply circuit including: A DC input terminal, wherein the DC input terminal is used to receive an external DC input power supply; The DC-DC buck converter has its input terminal connected to the DC input terminal, its output terminal used to output the operating voltage, and its input and output terminals grounded through multiple filter capacitors respectively. A voltage divider network is provided, with its lower end grounded. The output and feedback terminals of the DC-DC buck converter are connected through the voltage divider network.

[0009] Furthermore, the multi-channel DAC chip is a DAC82002DRXR chip, the voltage input terminal of the DAC82002DRXR chip is connected to the operating voltage, and the voltage input terminal of the DAC82002DRXR chip is grounded through a filter capacitor with a capacitance of 0.05 to 0.2 μF.

[0010] Furthermore, the microcontroller is an STM32F030F4P6 chip. The NRST interface of the STM32F030F4P6 chip is connected to the operating voltage via a pull-up resistor, and the NRST interface is grounded via a filter capacitor with a capacitance of 10~100 nF.

[0011] Furthermore, the communication driving circuit includes: A connection terminal, wherein the connection terminal includes two differential signal terminals; An RS485 transceiver, wherein the two differential signal terminals of the RS485 transceiver are connected to the two differential signal terminals of the connection terminal; the RS485 transceiver is also connected to the microcontroller via a transmit line and a receive line respectively.

[0012] Furthermore, the RS485 transceiver is a COS13487EESA chip.

[0013] Furthermore, the output terminals include at least a voltage interface of 0 to 10 V and a current interface of 0 to 20 mA.

[0014] On the other hand, according to embodiments of the present invention, an edge computing method with analog output control is provided, which uses an adaptive PID algorithm based on real-time sensor signal data to control various industrial equipment through the edge computing system described above, including: The edge computing system provides feedforward compensation when responding based on the inertial characteristics of the controlled object; The various industrial devices include at least one nonlinear actuator; the edge computing system maps or linearizes the adaptive PID output based on the nonlinear characteristics of the nonlinear actuator and drives the actuator in an analog manner. When the edge computing system receives a tracking command, it limits the rate of change of the output quantity according to the control setting value in the tracking command, and dynamically updates the control setting value according to the real-time sensor signal data.

[0015] The technical solutions provided by the embodiments of this disclosure may include the following beneficial effects: (a) It achieves hardware integration of edge computing and analog output, reducing system latency; (b) Design a modular hardware architecture to support flexible expansion of analog signal channels and communication protocols; (c) The analog output circuit and algorithm were optimized, which improved the accuracy and anti-interference ability of the output signal.

[0016] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of a module of an edge computing system provided in one embodiment; Figure 2 This is a circuit diagram of the control output module in an edge computing system provided in one embodiment; Figure 3 This is a schematic diagram of the control output module in one embodiment; Figure 4 This is a schematic diagram illustrating a scenario where an edge computing system is applied in practice, as shown in one embodiment. Detailed Implementation

[0019] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0020] Edge computing, as a device that brings computing and data storage closer to the data source, has wide applications in many fields such as industrial automation, smart buildings, and the Internet of Things. The technical solution provided in this application aims to solve related problems in analog output control of edge computing devices, and is suitable for industrial field data acquisition, real-time analysis, and actuator control scenarios. Through the technical solution of this application, high-precision, fast-response analog output control can be achieved, while improving compatibility between devices and reducing the difficulty and cost of system integration and deployment.

[0021] See Figures 1-4In one embodiment, an edge computing system with analog output control is provided, characterized in that the edge computing system includes: A data acquisition module, wherein the data acquisition module receives one or more sensor signals through an industrial protocol interface and preprocesses the acquired sensor signals to obtain sampled data; An edge computing module executes a control algorithm based on the sampled data to generate a target control quantity; the target control quantity is in the form of a digital signal. A control output module includes a communication driver circuit, a microcontroller connected to the communication driver circuit, a digital-to-analog converter circuit connected to the microcontroller, and an output circuit connected to the digital-to-analog converter circuit. The microcontroller receives digital signals from the edge computing module via the communication driver circuit. Under the control of the microcontroller, the digital-to-analog converter circuit converts the received digital signals into analog signals, which are then output via the output circuit. The control output module can convert the received target control quantity into a control signal and output it. The control signal is in analog signal form. The data acquisition module, the edge computing module, and the control output module are connected and communicate via a data interface, and the control output module and the edge computing module are integrated on the same hardware platform to reduce latency and improve efficiency. Preferably, the data acquisition module, the edge computing module, and the control output module are integrated on the same hardware platform. This edge computing system can output analog control signals and is easy to operate and maintain. Compared to traditional solutions using PLCs or external DAC devices, it significantly reduces the difficulty and cost of system integration.

[0022] See Figure 2 In one embodiment, the digital-to-analog converter circuit includes a multi-channel DAC chip, and the output circuit includes an output terminal and a multi-channel operational amplifier filter circuit connected to the output terminal. The operational amplifier filter circuit is connected to the multi-channel DAC chip, and the number of operational amplifier filter circuits is less than or equal to the number of output channels of the multi-channel DAC chip. The microcontroller can control the digital-to-analog converter circuit to output multiple analog signals.

[0023] The described edge computing system features multiple types of analog output interfaces and supports various communication protocols. The standardized interface design and rich communication protocol support ensure good compatibility with other devices from different manufacturers. When integrating multiple edge computing devices into a single system, complex interface conversions and protocol adaptations are unnecessary.

[0024] See Figure 2 , Figure 3In one embodiment, any one of the operational amplifier filter circuits includes an input resistor, a feedback resistor, and an isolation resistor, and the operational amplifier filter circuit further includes: An operational amplifier is provided, with its non-inverting input terminal connected to the output terminal of the corresponding channel of the multi-channel DAC chip via the input resistor; its inverting input terminal is grounded via a first resistor; and its output terminal is connected to the inverting input terminal via the feedback resistor. The output terminal is connected to the output terminal of the operational amplifier via the isolation resistor. The resistance of the feedback resistor is 1.5 to 3.5 times the resistance of the input resistor. See Figure 2 , Figure 3 In one embodiment, the control output module further includes a power supply circuit for supplying power to the circuits in the control output module, the power supply circuit including: A DC input terminal, wherein the DC input terminal is used to receive an external DC input power supply; The DC-DC buck converter has its input terminal connected to the DC input terminal, its output terminal used to output the operating voltage, and its input and output terminals grounded through multiple filter capacitors respectively. A voltage divider network is provided, with its lower end grounded. The output and feedback terminals of the DC-DC buck converter are connected through the voltage divider network.

[0025] See Figure 2 and Figure 3 In one embodiment, the multi-channel DAC chip is a DAC82002DRXR chip. The voltage input terminal of the DAC82002DRXR chip is connected to the operating voltage, and the voltage input terminal of the DAC82002DRXR chip is grounded through a filter capacitor with a capacitance of 0.05 to 0.2 μF. Compared with the lower precision digital-to-analog converter modules in the prior art, the 24-bit resolution digital-to-analog converter in this embodiment can more accurately convert digital signals into analog signals, thereby achieving high-precision analog output control. In industrial automated production, it can control the control error of analog parameters such as temperature and pressure within a very small range, improving product quality and operating efficiency.

[0026] See Figure 2 and Figure 3In one embodiment, the microcontroller is an STM32F030F4P6 chip. The NRST interface of the STM32F030F4P6 chip is connected to the operating voltage via a pull-up resistor, and the NRST interface is grounded via a filter capacitor with a capacitance of 10nF to 100nF.

[0027] See Figure 2 In one embodiment, the communication driving circuit includes: A connection terminal, wherein the connection terminal includes two differential signal terminals; An RS485 transceiver, wherein the two differential signal terminals of the RS485 transceiver are connected to the two differential signal terminals of the connection terminal; the RS485 transceiver is also connected to the microcontroller via a transmit line and a receive line respectively.

[0028] The RS485 to AO module is essentially an embedded digital-to-analog converter system with communication capabilities. It receives digital commands via an RS485 interface, which are then parsed by the main control MCU. The MCU then controls a DAC chip to perform digital-to-analog conversion, and finally, after signal conditioning circuitry (especially the VI conversion), a stable, interference-resistant industrial standard analog signal is generated, enabling precise and continuous control of field devices. Its core technology lies in a high-precision DAC, a stable VI conversion circuit, and reliable industrial communication protocol parsing.

[0029] See Figure 2 In one embodiment, the RS485 transceiver is a COS13487EESA chip.

[0030] See Figure 2 and Figure 3 In one embodiment, the output terminal includes at least a voltage interface of 0 to 10 V and a current interface of 0 to 20 mA.

[0031] See Figure 2In the control output module, the output circuit consists of two GS8552-SR operational amplifiers, U1.1 and U1.2. Input resistor R6, feedback resistor R11, and voltage divider resistor R8 are connected to U1.1, while input resistor R7, feedback resistor R10, and voltage divider resistor R9 are connected to U1.2, forming two outputs. These resistors protect the operational amplifiers and the multi-channel DAC chip U5 from damage caused by reverse current from the output terminal CN2. These two outputs are connected to pins 2 and 3 of the output terminal CN2. The output terminal CN2 is a WJ15EDGRC-3.81-4P. U1.1 and U1.2 are connected to the VOUTA and VOUTB output ports of the multi-channel DAC chip U5 (model U5DAC82002DRXR), respectively. The power supply interface VDD of the multi-channel DAC chip U5 is connected to the operating voltage VCC and is also grounded through a 0.1 μF capacitor C8 to provide filtering and protection.

[0032] The multi-channel DAC chip U5 has multiple interfaces connected to the microcontroller U4; the microcontroller U4 is an STM32F030F4P6, whose interfaces NRST, VDD, and VDDA are connected to the operating voltage VCC, and interfaces BOOT0 and VSS are grounded. Resistors R4 and R5, capacitors C5 and C7, etc., serve to stabilize voltage, filter, protect, isolate, and regulate the voltage level.

[0033] RS485 transceiver U3 uses a COS13487EESA chip and communicates with the edge computing module via connector CN1, model WAFER-PH2.0-4PZZ. Capacitor C8 stabilizes the operating voltage input to RS485 transceiver U3. Resistor R12 provides isolation and protection for microcontroller U4.

[0034] The DC input terminal receives an external DC input power supply VIN; VIN controls the switching of the DC-DC buck converter U2 via the protection resistor R1, and also serves as the input power supply for the entire module. U2 outputs a working voltage VCC, and its output terminal is connected to filter and voltage-stabilizing capacitors C1 and C2. The input terminal is connected to filter and voltage-stabilizing capacitors C3 and C4. Resistors R2 and R3 form a voltage divider network, with the lower end of the voltage divider network grounded. The OUTPUT and FEEDBACK ports of the DC-DC buck converter U2 are connected through this voltage divider network.

[0035] Figure 2 The specific embodiments shown can achieve Figure 3 The diagram shows two analog control signal outputs; however, it is clearly not limited to two outputs. Since the DAC section shown supports up to 16 channels, more channels can be output by increasing the number of output lines.

[0036] By integrating the DAC module with the edge calculator, multi-channel support is achieved. A high-precision DAC control module is used, incorporating RC filtering and TVS diodes to suppress surges, improving the device's stability and accuracy. The edge calculator communicates with the MCU controller via an RS485 interface using the Modbus protocol. The MCU controls multiple analog outputs by controlling the multi-channel DAC chip. Users write the required voltage or current values ​​into the corresponding registers; for example, 0x40011 controls the first analog output, 0x40012 controls the second, and so on, supporting up to 16 outputs. The MCU controller converts and amplifies the values ​​in the registers using a D / A converter to produce industry-standard analog signals. (Digital 0 is mapped to 0V, and digital 4095 is mapped to the full scale (e.g., 2.5V, 5V, or 10V reference voltage). The specific range selection is determined by the value of register 0x40099: range 1 is 2.5V; range 2 is 5V; range 3 is 10V).

[0037] When output current is required, the configuration register 0x40100 is set to 1, using a voltage-to-current (VI) converter. It utilizes operational amplifier and transistor feedback to ensure the final output current I is exactly equal to V / R (where V is the DAC output voltage and R is a precision setting resistor). (The value 0 in register 0x40011 maps to 0mA, and the value 4095 maps to the full-scale range of 20mA.) This embodiment demonstrates the setup of an RC filter, designed to remove high-frequency noise from the DAC output signal. This noise primarily originates from the switched capacitors within the DAC and high-frequency interference from the digital circuitry via power supply or coupling. Filtering is achieved at the analog voltage output of the DAC chip using a filter capacitor.

[0038] This embodiment also employs a TVS diode to suppress surges, protecting the sensitive DAC output circuitry from externally introduced transient high-voltage surges and electrostatic discharge (ESD) damage. These surges may originate from power fluctuations, lightning strikes, hot-plugging, static electricity, etc. A unidirectional TVS diode is selected here, with a cutoff voltage higher than the normal operating voltage of the signal.

[0039] In one embodiment, the edge control method employs hybrid control logic, deploying measures at the edge to predict load change trends and adjust outputs in advance. In the illustrated edge control method, multiple devices coordinate control; state estimation combining data from various sensors (such as temperature, pressure, and vibration) refers to controlling the analog output based on the location information collected by the edge calculator, i.e., the data from each sensor. For example, controlling the valve's opening position based on the current temperature value. With multiple sensors, to avoid the inaccuracy of a single sensor value affecting the actual control effect, energy saving is achieved while ensuring the expected requirements are met, and the anti-interference capability of the control signal is enhanced.

[0040] Reducing reliance on the cloud and enabling real-time analysis and control at the edge can meet the needs of applications with high real-time requirements, such as robot control and monitoring and control of high-speed moving objects.

[0041] See Figure 4 In one embodiment, the application scenario of the edge computing system is as follows: the edge computing device is connected to flow meters, frequency converters, and temperature, humidity, and pressure sensors, etc., and the edge end and the real-time values ​​of the sensors achieve constant pressure water supply, thereby saving energy and water. The role of the data acquisition layer is to enable various sensors (temperature and humidity meters, pressure gauges, electromagnetic flow meters, etc.) to be connected to the edge computing device through industrial protocols. The edge computing layer completes data acquisition → data processing → collaborative control. After the digital signal is converted by DAC, the AO control layer realizes the operation of frequency converters / valves / actuators, and controls the execution equipment layer (motor sets, regulating valves, hydraulic actuators, etc.).

[0042] See Figures 1-4 In one embodiment, an edge computing method with analog output control is provided, characterized in that the edge computing method uses an adaptive PID algorithm based on real-time sensor signal data to control various industrial devices through the edge computing system described above, including: The edge computing system provides feedforward compensation when responding based on the inertial characteristics of the controlled object; The various industrial devices include at least one nonlinear actuator; the edge computing system maps or linearizes the adaptive PID output based on the nonlinear characteristics of the nonlinear actuator and drives the actuator in an analog manner. When the edge computing system receives a tracking command, it limits the rate of change of the output quantity according to the control setting value in the tracking command, and dynamically updates the control setting value according to the real-time sensor signal data.

[0043] The following is a further explanation using a specific embodiment: Real-time sensors include temperature, humidity, CO2 concentration, light intensity sensors, and people counting. Building thermal inertia feedforward compensation is introduced (building envelopes have thermal inertia—even if the air supply stops, the room temperature will not immediately rise). The edge computing system continuously analyzes supply air temperature, return air temperature, current number of people in the environment, CO2 concentration, and external environmental data (temperature, light intensity), and dynamically controls the valve position via analog output. The valve characteristics are non-linear; the rate of change in flow rate due to changes in opening degree differs between 10% and 50%.

[0044] For example: Monday afternoon, cloudy turning sunny, peak electricity prices. External environmental sensors detect a sudden increase in solar radiation, and personnel statistics show high population density. The temperature setpoint is dynamically increased from 24.0°C to 25.2°C because calculations (see code 1) indicate a significant reduction in cooling energy consumption, resulting in the lowest total cost during peak electricity prices. The edge calculator receives the new setpoint of 25.2°C. Since the setpoint change is "upward," and the system recognizes this as a setpoint tracking command rather than a disturbance rejection, a smooth transition strategy is adopted. The AO control slowly reduces the air valve opening, allowing the temperature to smoothly transition to the new setpoint (see code 2), avoiding sudden changes in airflow that could cause discomfort to personnel.

[0045] Code 1: import numpy as np class SetpointOptimizer: def __init__(self): # Comfort Model Parameters (Simplified PMV) self.metabolic_rate = 1.2 # met, office activities self.clothing_insulation = 0.5 # clo, summer clothing self.air_velocity = 0.1# m / s # Economic model parameters self.electricity_price_peak = 1.5# Peak electricity price (yuan / kWh) self.electricity_price_normal = 0.6 # On-premium electricity price self.cooling_power_per_degree = 10.0 # Estimated increase in power consumption (kW) per 1°C decrease # Comfort Constraints self.max_acceptable_ppd = 20.0 # Maximum acceptable dissatisfaction rate (%) def calculate_dynamic_setpoint(self, sensor_data, current_setpoint=24.0): """ Calculate the dynamic setpoint based on real-time sensor data. sensor_data: A dictionary containing all sensor data. """ # 1. Unpack sensor data outdoor_temp = sensor_data['outdoor_temp'] indoor_temp = sensor_data['indoor_temp'] # Current indoor temperature indoor_humidity = sensor_data['indoor_humidity'] light_intensity = sensor_data['light_intensity'] occupant_count = sensor_data['occupant_count'] co2_level = sensor_data['co2_level'] is_peak_electricity = sensor_data['is_peak_electricity'] # 2. Calculate the current comfort level (simplified PMV-PPD model) # This is a simplified calculation; practical applications require the full PMV model or lookup table. estimated_pmv = self.estimate_pmv(indoor_temp, indoor_humidity, light_intensity, occupant_count) current_ppd = self.pmv_to_ppd(estimated_pmv) #3. Calculating thermal inertia compensation capability (core creative part) # Building thermal inertia allows us to appropriately increase the setpoint for a short period of time without immediately causing discomfort. thermal_inertia_buffer = self.calculate_thermal_inertia_buffer( indoor_temp, outdoor_temp, occupant_count, light_intensity ) # 4. Economic Calculation: How much cost can be saved by increasing the set value? energy_saving = self.calculate_energy_saving( current_setpoint, is_peak_electricity, thermal_inertia_buffer ) # 5. Multi-objective optimization decision-making: Finding a balance between comfort and economy new_setpoint = self.multi_objective_optimization( current_setpoint, current_ppd, energy_saving, thermal_inertia_buffer, is_peak_electricity ) return new_setpoint def estimate_pmv(self, temp, humidity, light, occupants): """Estimated Predicted Average Votes Value (PMV) - Extremely Simplified Version""" In practice, the standard PMV equation should be used; a linear model is used here for illustration. base_pmv = (temp - 24.0) 0.3# Temperature effect base_pmv += (humidity - 50) 0.01# Humidity effect base_pmv += (light - 500) 0.001# Light Influence (lux) base_pmv += max(0, occupants - 20) 0.05# Impact of personnel density return np.clip(base_pmv, -3, 3) # Limit to a reasonable range def pmv_to_ppd(self, pmv): Convert PMV to Predicted Percentage of Dissatisfaction (PPD) ppd = 100 - 95 np.exp(-0.03353 pmv 4 - 0.2179 pmv 2) return ppd def calculate_thermal_inertia_buffer(self, indoor_temp, outdoor_temp, occupants, light_intensity): """ Calculate the buffering capacity provided by building thermal inertia Return: Maximum safe increase of the set value (°C) """ # Thermal inertia buffer is based on: # - Indoor and outdoor temperature difference (smaller temperature difference, larger buffer) # - Current internal heat load (personnel, equipment) # - Building structural properties (simplified using fixed values) temp_difference = outdoor_temp - indoor_temp internal_load = occupants 100+ light_intensity 0.1# Estimate internal heat load # Core Formula: The smaller the temperature difference and the lower the internal load, the stronger the thermal inertia buffering capacity. base_buffer = 2.0 # Base buffer capacity (°C) buffer_reduction = temp_difference 0.1 + internal_load 0.001 available_buffer = max(0.5, base_buffer - buffer_reduction) # Minimum buffer size of 0.5°C return available_buffer def calculate_energy_saving(self, current_setpoint, is_peak_electricity, thermal_buffer): "Calculate the energy-saving benefits of increasing the set value." price = self.electricity_price_peak if is_peak_electricity elseself.electricity_price_normal # Available setting adjustment space = thermal inertia buffer potential_setpoint_increase = min(thermal_buffer, 3.0) # Maximum increase of 3°C # Energy saving estimate: Energy saved per 1°C increase × Electricity price energy_saving = (potential_setpoint_increase

[0046] self.cooling_power_per_degree price) return energy_saving def multi_objective_optimization(self, current_setpoint, current_ppd, energy_saving, thermal_buffer, is_peak_electricity): "Multi-objective optimization: balancing comfort and economic benefits" # Basic condition check: If the current comfort level is already very poor, the setting value should not be increased further. if current_ppd>self.max_acceptable_ppd 0.8:# The current PPD exceeds 80% of the maximum value. return current_setpoint # Keep the original setting value # Sacrificing comfort for energy saving is only considered during peak electricity price periods. if not is_peak_electricity: return current_setpoint# Prioritize comfort during off-peak hours # Decision Logic: Calculate the new setpoint based on energy-saving benefits and available buffer. saving_weight = energy_saving / 10.0 # Energy saving benefit weight buffer_utilization = 0.3 + saving_weight 0.7# Determines how much buffering capacity to use (30%-100%) setpoint_increase = thermal_buffer np.clip(buffer_utilization, 0, 1) new_setpoint = current_setpoint + setpoint_increase # Ensure that the increase in the setpoint is reasonable: it does not exceed the maximum allowable value and is a small increment. max_setpoint = 26.0 # Absolute maximum setpoint new_setpoint = min(new_setpoint, max_setpoint) # Ensure changes are smooth: each adjustment should not exceed 1.2°C. max_single_adjustment = 1.2 if new_setpoint - current_setpoint>max_single_adjustment: new_setpoint = current_setpoint + max_single_adjustment print(f"Optimization decision: {current_setpoint:.1f}°C -> {new_setpoint:.1f}°C | ") f"Energy saving: {energy_saving:.2f} | Thermal inertia buffer: {thermal_buffer:.1f}°C") return new_setpoint optimizer = SetpointOptimizer() sensor_data = { 'outdoor_temp': 35.0,# Outdoor temperature rises 'indoor_temp': 23.8, 'indoor_humidity': 55,# Humidity 'light_intensity': 800, # Enhanced light intensity (cloudy to sunny) 'occupant_count': 25,# Densely populated area 'co2_level': 800, # CO2 concentration 'is_peak_electricity': True # Peak electricity price } new_setpoint = optimizer.calculate_dynamic_setpoint(sensor_data,current_setpoint=24.0) print(f"New temperature setting: {new_setpoint}°C") Code 2: / / Global state variables float current_setpoint = 24.0f; / / Current actual setpoint (on the ramp) float target_setpoint = 24.0f; / / Target setpoint (e.g., 25.2) boolsetpoint_changed = false; / / PID status float integral = 0.0f; float prev_error = 0.0f; / / Valve nonlinearity lookup table (opening % -> flow coefficient) const float valve_flow_lut

[11] = { 0.0, 0.1, 0.2, 0.35, 0.5, 0.65, 0.75, 0.83, 0.9, 0.95, 1.0 }; / / Corresponds to 0%, 10%, ..., 100% / / Convert flow demand into valve opening (lookup table + interpolation) float flow_to_valve_opening(float flow_ratio) { if (flow_ratio<= 0.0f) return 0.0f; if (flow_ratio>= 1.0f) return 100.0f; for (int i = 0; i<10; i++) { if (flow_ratio<= valve_flow_lut[i+1]) { float t = (flow_ratio - valve_flow_lut[i]) / (valve_flow_lut[i+1] - valve_flow_lut[i]); return i 10.0f + t 10.0f; } } return 100.0f; } / / Main control loop (called every 100ms) void hvac_control_loop(float room_temp, float supply_temp) { / / 1. Set value ramp (smooth transition) if (setpoint_changed) { float delta = target_setpoint - current_setpoint; / / Ramp rate: 0.2°C per minute → 0.00033°C per 100ms float ramp_step = 0.00033f; if (fabsf(delta) <ramp_step) { current_setpoint = target_setpoint; setpoint_changed = false; } else { current_setpoint += (delta>0 ? ramp_step : -ramp_step); } } / / 2. Calculate the required cooling capacity (simplified) float temp_error = current_setpoint - room_temp; / / Note: If the setpoint is greater than the room temperature, cooling needs to be reduced. / / If temp_error > 0, it means the room temperature is lower than the set value, and the cooling fan should be reduced. / / 3. PID Calculation (Proportional + Integral Only, Derivative PID is Prone to Noise) float Kp = 0.8f, Ki = 0.02f; integral += temp_error 0.1f; / / dt=0.1s / / Anti-integral saturation (valve 0~100%) float flow_demand = Kp temp_error + Ki integral? flow_demand = fmaxf(0.0f, fminf(1.0f, flow_demand)); / / 4. Nonlinear compensation: Flow rate → Valve opening float valve_opening_pct = flow_to_valve_opening(flow_demand); / / 5. Output to AO (0~10V or 4~20mA) float ao_voltage = 0.1f valve_opening_pct; / / 0~100% → 0~10V write_DAC_channel(AO_CHANNEL, ao_voltage); / / Send to analog output module } Local edge computing devices and DACs work together to effectively reduce reliance on the cloud, achieving low-latency real-time control and overcoming the shortcomings of existing technologies in terms of latency and cloud dependence. The system employs a modular hardware design and dynamically configurable protocol software, offering excellent scalability. Channels can be expanded as needed without replacing the edge computing device interface (traditional devices typically require circuit redesign). It also provides multiple analog output interfaces, including voltage output (e.g., 0~10 V) and current output (e.g., 4~20 mA). An adaptive PID algorithm based on real-time data provides adaptive control capabilities, with control parameters dynamically optimized to adapt to environmental changes (offering greater adaptability compared to traditional solutions relying on fixed parameters). Furthermore, the standardized interface design further enhances system compatibility.

[0047] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the indirect coupling or communication connection of the apparatus or modules can be electrical, mechanical, or other forms.

[0048] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0049] Furthermore, the functional modules in the various embodiments of this application can be integrated into one processing unit, or each module can exist physically separately, or two or more modules can be integrated into one unit. The integrated unit can be implemented in hardware or in a combination of hardware and software functional units.

[0050] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

Claims

1. An edge computing system with analog output control, characterized in that, The edge computing system includes: A data acquisition module, wherein the data acquisition module receives one or more sensor signals through an industrial protocol interface and preprocesses the acquired sensor signals to obtain sampled data; An edge computing module executes a control algorithm based on the sampled data to generate a target control quantity; the target control quantity is in the form of a digital signal. A control output module includes a communication driver circuit, a microcontroller connected to the communication driver circuit, a digital-to-analog converter circuit connected to the microcontroller, and an output circuit connected to the digital-to-analog converter circuit. The microcontroller receives digital signals from the edge computing module via the communication driver circuit. Under the control of the microcontroller, the digital-to-analog converter circuit converts the received digital signals into analog signals, which are then output via the output circuit. The control output module can convert the received target control quantity into a control signal and output it. The control signal is in analog signal form. The data acquisition module, the edge computing module, and the control output module are connected and communicate with each other through a data interface, and the control output module and the edge computing module are integrated on the same hardware platform.

2. The edge computing system as described in claim 1, characterized in that, The digital-to-analog converter circuit includes a multi-channel DAC chip, and the output circuit includes an output terminal and a multi-channel operational amplifier filter circuit connected to the output terminal. The operational amplifier filter circuit is connected to the multi-channel DAC chip, and the number of operational amplifier filter circuits is less than or equal to the number of output channels of the multi-channel DAC chip. The microcontroller can control the digital-to-analog converter circuit to output multiple analog signals.

3. The edge computing system as described in claim 2, characterized in that, Any of the operational amplifier filter circuits includes an input resistor, a feedback resistor, and an isolation resistor. The operational amplifier filter circuit also includes: An operational amplifier is provided, with its non-inverting input terminal connected to the output terminal of the corresponding channel of the multi-channel DAC chip via the input resistor; its inverting input terminal is grounded via a first resistor; and its output terminal is connected to the inverting input terminal via the feedback resistor. The output terminal is connected to the output terminal of the operational amplifier via the isolation resistor; The resistance of the feedback resistor is 1.5 to 3.5 times the resistance of the input resistor.

4. The edge computing system as described in claim 3, characterized in that, The control output module further includes a power supply circuit for supplying power to each circuit in the control output module, the power supply circuit including: A DC input terminal, wherein the DC input terminal is used to receive an external DC input power supply; The DC-DC buck converter has its input terminal connected to the DC input terminal, its output terminal used to output the operating voltage, and its input and output terminals grounded through multiple filter capacitors respectively. A voltage divider network is provided, with its lower end grounded. The output and feedback terminals of the DC-DC buck converter are connected through the voltage divider network.

5. The edge computing system as described in claim 4, characterized in that, The multi-channel DAC chip is a DAC82002DRXR chip. The voltage input terminal of the DAC82002DRXR chip is connected to the operating voltage, and the voltage input terminal of the DAC82002DRXR chip is grounded through a filter capacitor with a capacitance of 0.05 to 0.2 μF.

6. The edge computing system as described in claim 4, characterized in that, The microcontroller is an STM32F030F4P6 chip. The NRST interface of the STM32F030F4P6 chip is connected to the operating voltage via a pull-up resistor, and the NRST interface is grounded via a filter capacitor with a capacitance of 10~100nF.

7. The edge computing system as described in claim 2, characterized in that, The communication driving circuit includes: A connection terminal, wherein the connection terminal includes two differential signal terminals; An RS485 transceiver, wherein the two differential signal terminals of the RS485 transceiver are connected to the two differential signal terminals of the connection terminal; the RS485 transceiver is also connected to the microcontroller via a transmit line and a receive line respectively.

8. The edge computing system as described in claim 7, characterized in that, The RS485 transceiver is a COS13487EESA chip.

9. The edge computing system as described in claim 2, characterized in that, The output terminals include at least a voltage interface of 0 to 10 V and a current interface of 0 to 20 mA.

10. An edge computing method with analog output control, characterized in that, The edge computing method, through the edge computing system as described in any one of claims 1 to 9, employs an adaptive PID algorithm based on real-time sensor signal data to control various industrial equipment, including: The edge computing system provides feedforward compensation when responding based on the inertial characteristics of the controlled object; The various industrial devices include at least one nonlinear actuator; the edge computing system maps or linearizes the adaptive PID output based on the nonlinear characteristics of the nonlinear actuator and drives the actuator in an analog manner. When the edge computing system receives a tracking command, it limits the rate of change of the output quantity according to the control setting value in the tracking command, and dynamically updates the control setting value according to the real-time sensor signal data.

Images