Calculation method of array lateral electrode spacing

Abstract

The invention relates to the technical field of array calculation, in particular to an array lateral electrode spacing calculation method, which comprises the following steps of: determining a working mode of an array calculation unit, acquiring an initial electrode spacing, adjusting an electrode spacing value based on an optimization parameter, and calculating an actual electrode spacing. The electrode spacing is dynamically optimized by integrating multiple factors such as signal transmission efficiency, energy consumption distribution and heat distribution, and the system performance and stability are improved. The problems of signal crosstalk and energy consumption in a high-density calculation scene can be effectively solved, the complex calculation requirement is met, and the adaptability and the overall performance of the array calculation unit are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of electrode design and optimization, and in particular to a method for calculating the lateral electrode spacing of an array. Background Art

[0002] With the continuous development of array computing technology, the design and optimization of electrode spacing plays a key role in improving computing efficiency, reducing energy consumption, and enhancing system stability. However, existing methods for calculating the lateral electrode spacing in arrays still have some shortcomings in practical applications, especially in terms of accuracy and adaptability in complex computing scenarios.

[0003] After searching, a computing device and computing method with publication number CN114065124B was disclosed, and the publication date is March 18, 2025. This patent sets a data preprocessing unit before the computing unit array, and generates a control signal according to the data of the input matrix to indicate the working status of the computing unit, thereby effectively reducing the energy consumption of the control unit during the matrix operation and improving the computing speed. However, this technical solution does not involve a specific calculation method for the lateral electrode spacing of the array, and fails to fully consider the impact of the electrode spacing on the computing performance. In addition, it mainly focuses on the improvement of energy consumption and speed, and the potential relationship between the electrode spacing design and signal transmission efficiency and system stability is insufficiently explored, which may cause signal crosstalk or increased energy loss in high-density array computing scenarios.

[0004] After searching, a computing system and computing method with publication number CN116504285B was disclosed, and the publication date is October 18, 2024. This patent solves the coupling problem between the input circuit and the array by constructing a method in which the data in the memory bits connected to the corresponding input word lines are summed as a constant, and realizes the decoupling of subsystems in complex analog circuit systems, thereby improving the reliability and efficiency of the system. However, this technical solution mainly focuses on the data coupling problem between memory bits and does not directly involve the optimization design of the lateral electrode spacing of the array. Due to the lack of in-depth analysis of the relationship between electrode spacing and computing performance, the calculation accuracy and the overall performance of the system may be affected by unreasonable electrode spacing design in high-precision computing tasks.

[0005] The above issues demonstrate that existing computing devices and methods have limitations in accurately calculating and optimizing the lateral electrode spacing of an array. In particular, their adaptability in high-density, high-precision computing scenarios needs to be further improved. Therefore, the present invention provides a method for calculating the lateral electrode spacing of an array. This method aims to improve signal transmission efficiency, reduce energy consumption, and enhance system stability by optimizing the electrode spacing design, thereby meeting the high-performance computing requirements of complex computing scenarios. Summary of the Invention

[0006] The present invention aims to address at least one of the technical problems existing in the related art. To this end, the present invention proposes a method for calculating the lateral electrode spacing of an array. By optimizing the electrode spacing design, the method meets the requirements of complex computing scenarios for signal transmission efficiency, energy consumption control, and system stability, while also improving the overall performance of array computing.

[0007] A first aspect of the present invention provides a method for calculating the electrode spacing in the lateral direction of an array, comprising: determining an operating mode of an array computing unit; obtaining an initial electrode spacing of the array computing unit based on the operating mode; adjusting the electrode spacing value of the array computing unit according to the initial electrode spacing and preset optimization parameters; and calculating the actual electrode spacing of the array computing unit based on the adjusted electrode spacing value.

[0008] According to the first aspect of the present invention, the method for calculating the electrode spacing in the lateral direction of the array provided by the embodiment of the present invention determines the operating mode of the array computing unit and can flexibly adjust the electrode spacing calculation method to meet the specific requirements of different operating modes. This helps improve the adaptability and performance of the array computing unit in different computing tasks. Obtaining the initial electrode spacing based on the operating mode can more accurately reflect the actual requirements of the array computing unit in the current operating state. By comprehensively considering multiple factors, a more accurate and reasonable electrode spacing value can be calculated, thereby improving signal transmission stability and reducing energy loss. Using the adjusted electrode spacing value to calculate the actual electrode spacing can achieve dynamic optimization of the electrode spacing in the array computing unit. This method can reduce signal crosstalk in high-density computing scenarios while enhancing the overall stability of the system. In addition, by smoothing the electrode spacing value, the continuity of signal transmission can be further improved and performance fluctuations caused by sudden changes in electrode spacing can be reduced. Therefore, the method for calculating the electrode spacing in the lateral direction of the array provided by the first aspect of the present invention, by comprehensively considering the operating mode, computing task characteristics, and electrode spacing optimization parameters, achieves accurate calculation and efficient adjustment of electrode spacing, thereby improving array computing performance, energy consumption control capabilities, and system stability.

[0009] According to one embodiment of the present invention, the working mode includes a high-density computing mode; based on the working mode of the array computing unit, the step of obtaining the initial electrode spacing of the array computing unit includes: collecting the input signal frequency and signal strength information of the array computing unit; determining the initial electrode spacing value of the array computing unit based on the input signal frequency and signal strength information; comparing the initial electrode spacing value with a preset minimum spacing threshold, and if the initial electrode spacing value is less than the minimum spacing threshold, then using the minimum spacing threshold as the initial electrode spacing value.

[0010] According to one embodiment of the present invention, the initial electrode spacing value, the minimum spacing threshold and the actual electrode spacing value satisfy the following relationship: Dactual=Dinitial+ΔDoptimized; wherein Dactual is the actual electrode spacing value, Dinitial is the initial electrode spacing value, and ΔDoptimized is the optimization parameter adjustment value.

[0011] According to one embodiment of the present invention, the working mode includes a low-power computing mode; based on the working mode of the array computing unit, the step of obtaining the initial electrode spacing of the array computing unit includes: collecting energy consumption distribution data and thermal distribution information of the array computing unit; determining the initial electrode spacing value of the array computing unit based on the energy consumption distribution data and the thermal distribution information; comparing the initial electrode spacing value with a preset maximum spacing threshold; if the initial electrode spacing value is greater than the maximum spacing threshold, then using the maximum spacing threshold as the initial electrode spacing value.

[0012] According to one embodiment of the present invention, the actual electrode spacing value satisfies the following relationship: Dactual=Dinitial×(1+α); wherein Dactual is the actual electrode spacing value, Dinitial is the initial electrode spacing value, α is the optimization coefficient, and 0≤α≤1.

[0013] According to one embodiment of the present invention, the step of calculating the actual electrode spacing of the array computing unit based on the adjusted electrode spacing value includes: determining the signal transmission efficiency and energy loss distribution of the array computing unit; setting the optimization coefficient α incrementally according to a fixed step size to obtain the corresponding signal transmission efficiency under different optimization coefficients; setting the optimization coefficient β incrementally according to a fixed step size to obtain the corresponding energy loss distribution under different optimization coefficients; setting the objective function, and when the objective function reaches the optimal value, obtaining the actual electrode spacing value.

[0014] According to one embodiment of the present invention, the step of calculating the actual electrode spacing of the array computing unit based on the adjusted electrode spacing value further includes: determining the working state of the array computing unit; adjusting the actual electrode spacing value based on the working state of the array computing unit; and / or, in the event that the actual electrode spacing value suddenly changes, increasing or decreasing the actual electrode spacing value according to a preset step size.

[0015] According to one embodiment of the present invention, the signal transmission efficiency and energy loss distribution of the array computing unit are obtained by: setting multiple monitoring points in the array computing unit and collecting the signal amplitude and energy loss value of each monitoring point in real time; constructing a signal transmission efficiency matrix and an energy loss distribution matrix based on the collected signal amplitude and energy loss values; and calculating the overall signal transmission efficiency and energy loss distribution of the array computing unit through matrix operations.

[0016] According to one embodiment of the present invention, the optimization coefficient α and the optimization coefficient β satisfy the following relationship: α + β = 1, where α is the signal transmission efficiency optimization coefficient and β is the energy loss optimization coefficient. The adjustment range of the optimization coefficient is determined by preset constraints to ensure that the array computing unit achieves a balance between signal transmission efficiency and energy loss.

[0017] According to one embodiment of the present invention, the objective function is expressed as: F = k1 × Esignal + k2 × Eenergy; where F is the objective function value, Esignal is the signal transmission efficiency, Eenergy is the energy loss distribution, k1 and k2 are weight coefficients, and k1 + k2 = 1. By adjusting the weight coefficients k1 and k2, signal transmission efficiency or energy loss distribution can be prioritized according to actual needs.

[0018] According to one embodiment of the present invention, the adjustment process of the actual electrode spacing value includes: setting multiple adjustment modules in the array computing unit, each adjustment module corresponding to an electrode spacing adjustment area; adjusting the electrode spacing of each adjustment module based on the actual electrode spacing value; during the adjustment process, using a piecewise linear interpolation method to smooth the electrode spacing value to avoid performance fluctuations caused by sudden changes in the electrode spacing.

[0019] According to one embodiment of the present invention, thermal distribution information of an array computing unit is obtained by: arranging multiple temperature sensors on the surface of the array computing unit and collecting temperature data from each temperature sensor in real time; constructing a thermal distribution map based on the collected temperature data; analyzing the heat concentration area of ​​the array computing unit through the thermal distribution map, and adjusting the electrode spacing value accordingly to reduce the temperature of the heat concentration area.

[0020] According to one embodiment of the present invention, the energy consumption distribution data of the array computing unit is obtained in the following manner: multiple energy consumption monitoring modules are set in the array computing unit to collect energy consumption data of each energy consumption monitoring module in real time; based on the collected energy consumption data, an energy consumption distribution map is constructed; the high energy consumption area of ​​the array computing unit is analyzed through the energy consumption distribution map, and the electrode spacing value is adjusted accordingly to reduce energy loss in the high energy consumption area.

[0021] According to one embodiment of the present invention, the signal strength information of the array computing unit is obtained in the following manner: a signal detection circuit is provided at the input end of the array computing unit to collect the amplitude and frequency of the input signal in real time; a signal strength value is calculated based on the collected signal amplitude and frequency; the characteristics of the input signal are analyzed by the signal strength value, and the electrode spacing value is adjusted accordingly to improve the signal transmission efficiency.

[0022] According to one embodiment of the present invention, the frequency of the input signal of the array computing unit is obtained in the following manner: a frequency detection module is set at the input end of the array computing unit to collect frequency data of the input signal in real time; based on the collected frequency data, an average frequency value of the input signal is calculated; the change trend of the input signal is analyzed by the average frequency value, and the electrode spacing value is adjusted accordingly to adapt to input signals of different frequencies.

[0023] Through the aforementioned specific technical means, the present invention's method for calculating the lateral electrode spacing of an array effectively addresses the shortcomings of existing technologies, particularly their poor adaptability in high-density, high-precision computing scenarios. By optimizing the electrode spacing design, this method significantly improves signal transmission efficiency, reduces energy consumption, and enhances system stability, thus meeting the high-performance computing requirements of complex computing scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 A schematic flow chart of a method for calculating the lateral electrode spacing of an array provided in an embodiment of the present invention.

[0025] In the figure: 1. Working mode determination module; 2. Initial electrode spacing acquisition module; 3. Optimization parameter adjustment module; 4. Actual electrode spacing calculation module. DETAILED DESCRIPTION

[0026] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0027] The present invention provides a method for calculating the electrode spacing on the side of an array. Figure 1 The flowchart shown is explained in detail. Figure 1 The complete process from determining the operating mode to calculating the actual electrode spacing is presented in Figure 1. The operating mode determination module 1, initial electrode spacing acquisition module 2, optimization parameter adjustment module 3, and actual electrode spacing calculation module 4, involved in this process, together constitute the core components of this method. The following describes the technical solution of the present invention in detail, combining the specific operating principles and coordination relationships of these modules.

[0028] In practical applications, array computing units typically need to adjust their electrode spacing to meet performance requirements based on different operating scenarios and task requirements. The operating mode of the array computing unit is first determined by operating mode determination module 1. This module collects input signal characteristic information related to the array computing unit and, based on this information, determines the current operating mode. For example, in high-density computing mode, operating mode determination module 1 collects input signal frequency and signal strength information from the array computing unit. This information is acquired in real time by a frequency detection module and signal detection circuit located at the input of the array computing unit. The frequency detection module collects input signal frequency data and calculates the average frequency value, while the signal detection circuit collects the input signal amplitude and frequency to calculate the signal strength value. By analyzing the frequency variation trend and signal strength characteristics of the input signal, operating mode determination module 1 can accurately determine whether the current operation is in high-density computing mode. Similarly, in low-power computing mode, operating mode determination module 1 collects energy consumption distribution data and thermal distribution information from the array computing unit. This information is acquired in real time by temperature sensors located on the surface of the array computing unit and an internal energy consumption monitoring module. The temperature sensor collects temperature data from various locations and constructs a heat distribution map to analyze heat concentration areas. The energy consumption monitoring module collects energy consumption data from each area and generates an energy consumption distribution map to identify high-energy consumption areas. Through comprehensive analysis of this information, the operating mode determination module 1 ultimately determines the current operating mode.

[0029] After determining the working mode, the initial electrode spacing acquisition module 2 obtains the initial electrode spacing value of the array computing unit according to the working mode. For the high-density computing mode, the initial electrode spacing acquisition module 2 calculates the initial electrode spacing value based on the input signal frequency and signal strength information, and compares it with the preset minimum spacing threshold. If the calculated initial electrode spacing value is less than the minimum spacing threshold, the minimum spacing threshold is used as the initial electrode spacing value. This processing method ensures that the electrode spacing is not too small to cause signal crosstalk or system instability. For the low-power computing mode, the initial electrode spacing acquisition module 2 calculates the initial electrode spacing value based on the energy consumption distribution data and thermal distribution information, and compares it with the preset maximum spacing threshold. If the calculated initial electrode spacing value is greater than the maximum spacing threshold, the maximum spacing threshold is used as the initial electrode spacing value. This processing method avoids the increase in energy loss or decrease in signal transmission efficiency caused by excessive electrode spacing. In this way, the initial electrode spacing acquisition module 2 can flexibly adjust the initial value of the electrode spacing according to the requirements of different working modes.

[0030] Next, the optimization parameter adjustment module 3 adjusts the electrode spacing value based on the initial electrode spacing value and preset optimization parameters. The core of the optimization parameter adjustment module 3 lies in the introduction of optimization coefficients α and β, which correspond to the signal transmission efficiency optimization coefficient and energy loss optimization coefficient, respectively. These two optimization coefficients satisfy the relationship α + β = 1, and their adjustment range is determined by preset constraints. The optimization parameter adjustment module 3 obtains the signal transmission efficiency and energy loss distribution under different optimization coefficients by incrementally setting the optimization coefficient α and decrementing the optimization coefficient β. Specifically, the optimization parameter adjustment module 3 sets multiple monitoring points in the array computing unit and collects the signal amplitude and energy loss values ​​at each monitoring point in real time. Based on the collected data, the optimization parameter adjustment module 3 constructs a signal transmission efficiency matrix and an energy loss distribution matrix, and calculates the overall signal transmission efficiency and energy loss distribution of the array computing unit through matrix operations. Based on this data, the optimization parameter adjustment module 3 sets the objective function F = k1 × Esignal + k2 × Eenergy, where k1 and k2 are weight coefficients, satisfying k1 + k2 = 1. By adjusting the weight coefficients k1 and k2, the signal transmission efficiency or energy loss distribution can be prioritized according to actual needs. When the objective function reaches the optimal value, the optimization parameter adjustment module 3 outputs the adjusted electrode spacing value.

[0031] Finally, the actual electrode spacing calculation module 4 calculates the actual electrode spacing of the array computing unit based on the adjusted electrode spacing value. The actual electrode spacing calculation module 4 uses the formula Dactual = Dinitial + ΔDOptimized or Dactual = Dinitial × (1 + α) for calculation, where Dactual is the actual electrode spacing value, Dinitial is the initial electrode spacing value, ΔDOptimized is the optimization parameter adjustment value, and α is the optimization coefficient. During the calculation process, the actual electrode spacing calculation module 4 also considers the operating status of the array computing unit. For example, when the actual electrode spacing value suddenly changes, the actual electrode spacing calculation module 4 will increment or decrement the actual electrode spacing value according to a preset step size to avoid performance fluctuations caused by sudden changes in electrode spacing. In addition, the actual electrode spacing calculation module 4 smoothes the electrode spacing value using piecewise linear interpolation to further improve signal transmission continuity. To achieve the above functions, the actual electrode spacing calculation module 4 sets multiple adjustment modules in the array computing unit, each adjustment module corresponding to an electrode spacing adjustment region. Based on the calculated actual electrode spacing value, the actual electrode spacing calculation module 4 adjusts the electrode spacing of each adjustment module separately, thereby achieving dynamic optimization of the electrode spacing.

[0032] Throughout the entire process, the various modules work closely together through data transmission and logic control. The working mode determination module 1 passes the determined working mode information to the initial electrode spacing acquisition module 2. The initial electrode spacing acquisition module 2 calculates the initial electrode spacing value according to the working mode and passes it to the optimization parameter adjustment module 3. The optimization parameter adjustment module 3 calculates the adjusted electrode spacing value based on the initial electrode spacing value and the optimization parameters, and passes the result to the actual electrode spacing calculation module 4. The actual electrode spacing calculation module 4 finally calculates the actual electrode spacing value and applies it to the electrode spacing adjustment of the array computing unit. Through this modular design and collaborative working mechanism, the array lateral electrode spacing calculation method provided by the present invention can effectively solve the shortcomings of the existing technology, especially the problem of poor adaptability in high-density, high-precision computing scenarios.

[0033] In order to better enable relevant personnel in this technical field to fully understand and implement the present invention, the specific implementation principle of the present invention is further supplemented below in combination with a specific application scenario.

[0034] In high-density computing scenarios, array computing units need to process high-frequency input signals while ensuring the stability of signal transmission. First, the working mode determination module 1 collects the frequency data of the input signal through the frequency detection module and calculates the average frequency value; at the same time, the signal detection circuit collects the amplitude and frequency of the input signal and calculates the signal strength value. Based on the above information, the working mode determination module 1 determines that it is currently in high-density computing mode. Subsequently, the initial electrode spacing acquisition module 2 calculates the initial electrode spacing value based on the input signal frequency and signal strength information. For example, if the input signal frequency is high and the signal strength is large, the initial electrode spacing value will tend to be smaller to meet the high-density computing requirements. However, in order to avoid signal crosstalk or system instability due to too small an electrode spacing, the initial electrode spacing acquisition module 2 compares the calculated initial electrode spacing value with a preset minimum spacing threshold. When the initial electrode spacing value is less than the minimum spacing threshold, the minimum spacing threshold is set to the initial electrode spacing value.

[0035] Next, the optimization parameter adjustment module 3 adjusts the electrode spacing value based on the initial electrode spacing value and preset optimization parameters. The optimization parameter adjustment module 3 introduces a signal transmission efficiency optimization coefficient α and an energy loss optimization coefficient β, which satisfy the relationship α + β = 1. The optimization parameter adjustment module 3 sets multiple monitoring points in the array computing unit and collects the signal amplitude and energy loss values ​​at each monitoring point in real time. Based on this data, it constructs a signal transmission efficiency matrix and an energy loss distribution matrix, and calculates the overall signal transmission efficiency and energy loss distribution of the array computing unit through matrix operations. Based on this, the optimization parameter adjustment module 3 sets the objective function F = k1 × E signal + k2 × E energy, where k1 and k2 are weight coefficients, respectively, and satisfy k1 + k2 = 1. By adjusting the weight coefficients k1 and k2, the signal transmission efficiency or energy loss distribution can be prioritized according to actual needs. For example, in high-density computing scenarios, the value of k1 can be appropriately increased to prioritize signal transmission efficiency. When the objective function reaches the optimal value, the optimization parameter adjustment module 3 outputs the adjusted electrode spacing value.

[0036] Finally, the actual electrode spacing calculation module 4 calculates the actual electrode spacing of the array computing unit based on the adjusted electrode spacing value. The actual electrode spacing calculation module 4 uses the formula Dactual=DInitial+ΔDOptimized for calculation. During the calculation process, the actual electrode spacing calculation module 4 will also dynamically adjust the electrode spacing value according to the working state of the array computing unit. For example, when the actual electrode spacing value suddenly changes, the actual electrode spacing calculation module 4 will increase or decrease the actual electrode spacing value according to a preset step size to avoid performance fluctuations caused by sudden changes in electrode spacing. In addition, the actual electrode spacing calculation module 4 also smoothes the electrode spacing value through piecewise linear interpolation, thereby improving the continuity of signal transmission. In order to achieve the above functions, the actual electrode spacing calculation module 4 sets multiple adjustment modules in the array computing unit, and each adjustment module corresponds to an electrode spacing adjustment area. Based on the calculated actual electrode spacing value, the actual electrode spacing calculation module 4 adjusts the electrode spacing of each adjustment module respectively, thereby achieving dynamic optimization of the electrode spacing.

[0037] In low-power computing scenarios, array computing units need to reduce energy consumption and control heat distribution. The working mode determination module 1 collects temperature data at each location through temperature sensors arranged on the surface of the array computing unit and generates a heat distribution map; at the same time, the energy consumption monitoring module collects energy consumption data from each area and generates an energy consumption distribution map. Based on the above information, the working mode determination module 1 determines that it is currently in low-power computing mode. Subsequently, the initial electrode spacing acquisition module 2 calculates the initial electrode spacing value based on the energy consumption distribution data and heat distribution information. For example, if the energy consumption of a certain area is high and the heat concentration is large, the initial electrode spacing value will tend to be a larger value to reduce the energy loss and temperature of the area. In order to avoid a decrease in signal transmission efficiency due to an excessively large electrode spacing, the initial electrode spacing acquisition module 2 compares the calculated initial electrode spacing value with a preset maximum spacing threshold. When the initial electrode spacing value is greater than the maximum spacing threshold, the maximum spacing threshold is set as the initial electrode spacing value.

[0038] Optimization parameter adjustment module 3 also adjusts the electrode spacing value based on the initial electrode spacing value and the preset optimization parameters. In low-power computing scenarios, optimization parameter adjustment module 3 can prioritize reducing energy loss distribution by appropriately increasing the value of k2. When the objective function reaches the optimal value, optimization parameter adjustment module 3 outputs the adjusted electrode spacing value.

[0039] The actual electrode spacing calculation module 4 calculates the actual electrode spacing of the array computing unit based on the adjusted electrode spacing value. The actual electrode spacing calculation module 4 uses the formula Dactual = Dinitial × (1 + α) for calculation. During the calculation process, the actual electrode spacing calculation module 4 will also dynamically adjust the electrode spacing value according to the working state of the array computing unit. For example, when the actual electrode spacing value suddenly changes, the actual electrode spacing calculation module 4 will increase or decrease the actual electrode spacing value according to a preset step size to avoid performance fluctuations caused by sudden changes in electrode spacing. In addition, the actual electrode spacing calculation module 4 also smoothes the electrode spacing value through piecewise linear interpolation, thereby improving the continuity of signal transmission. In order to achieve the above functions, the actual electrode spacing calculation module 4 sets multiple adjustment modules in the array computing unit, and each adjustment module corresponds to an electrode spacing adjustment area. Based on the calculated actual electrode spacing value, the actual electrode spacing calculation module 4 adjusts the electrode spacing of each adjustment module respectively, thereby achieving dynamic optimization of the electrode spacing.

[0040] Throughout the entire process, the various modules work closely together through data transmission and logic control. The working mode determination module 1 passes the determined working mode information to the initial electrode spacing acquisition module 2. The initial electrode spacing acquisition module 2 calculates the initial electrode spacing value according to the working mode and passes it to the optimization parameter adjustment module 3. The optimization parameter adjustment module 3 calculates the adjusted electrode spacing value based on the initial electrode spacing value and the optimization parameters, and passes the result to the actual electrode spacing calculation module 4. The actual electrode spacing calculation module 4 finally calculates the actual electrode spacing value and applies it to the electrode spacing adjustment of the array computing unit. Through this modular design and collaborative working mechanism, the array lateral electrode spacing calculation method provided by the present invention can effectively solve the shortcomings of the existing technology, especially the problem of poor adaptability in high-density, high-precision computing scenarios.

[0041] Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments, or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for calculating the electrode spacing in the lateral direction of an array, characterized in that: The following steps are involved: Determine the operating mode of the array computing unit; Based on the working mode, obtaining an initial electrode spacing value of the array computing unit; Adjusting the electrode spacing value of the array computing unit according to the initial electrode spacing value and the preset optimization parameters; Based on the adjusted electrode spacing value, the actual electrode spacing value of the array computing unit is calculated.

2. The method for calculating the lateral electrode spacing of an array according to claim 1, wherein: The working mode includes a high-density computing mode; The step of obtaining the initial electrode spacing value of the array computing unit based on the working mode includes: Collecting input signal frequency and signal strength information of the array computing unit; Determining an initial electrode spacing value of an array computing unit based on the input signal frequency and signal strength information; The initial electrode spacing value is compared with a preset minimum spacing threshold value. If the initial electrode spacing value is smaller than the minimum spacing threshold value, the minimum spacing threshold value is used as the initial electrode spacing value.

3. The method for calculating the lateral electrode spacing of an array according to claim 1, wherein: The actual electrode spacing value, the initial electrode spacing value and the optimized parameter adjustment value satisfy the following relationship: Dactual = Dinitial + ΔDoptimized; Where Dactual is the actual electrode spacing value, Dinitial is the initial electrode spacing value, and ΔDoptimize is the optimized parameter adjustment value.

4. The method for calculating the lateral electrode spacing of an array according to claim 1, wherein: The working mode includes a low-power computing mode; The step of obtaining the initial electrode spacing value of the array computing unit based on the working mode includes: Collect energy consumption distribution data and thermal distribution information of array computing units; Determining an initial electrode spacing value of an array computing unit based on the energy consumption distribution data and the heat distribution information; The initial electrode spacing value is compared with a preset maximum spacing threshold value. If the initial electrode spacing value is greater than the maximum spacing threshold value, the maximum spacing threshold value is used as the initial electrode spacing value.

5. The method for calculating the lateral electrode spacing of an array according to claim 1, wherein: The actual electrode spacing value satisfies the following relationship: Dactual = Dinitial × (1 + α); Wherein, Dactual is the actual electrode spacing value, Dinitial is the initial electrode spacing value, α is the optimization coefficient, and 0≤α≤1.

6. The method for calculating the lateral electrode spacing of an array according to claim 1, wherein: The step of calculating the actual electrode spacing value of the array computing unit based on the adjusted electrode spacing value comprises: Determine the signal transmission efficiency and energy loss distribution of array computing units; The optimization coefficient α is set incrementally according to a fixed step size to obtain the signal transmission efficiency under different optimization coefficients; The optimization coefficient β is set in a decreasing manner according to a fixed step size to obtain the energy loss distribution under different optimization coefficients; The objective function is set, and when the objective function reaches the optimal value, the actual electrode spacing is obtained.

7. The method for calculating the lateral electrode spacing of an array according to claim 6, characterized in that: The expression of the objective function is: F = k1 × E signal + k2 × E energy; Wherein, F is the objective function value, Esignal is the signal transmission efficiency, Eenergy is the energy loss distribution, k1 and k2 are weight coefficients, and k1+k2=1.

8. The method for calculating the lateral electrode spacing of an array according to claim 1, wherein: The step of calculating the actual electrode spacing value of the array computing unit based on the adjusted electrode spacing value further includes: Determine the working status of the array computing unit; Adjusting the actual electrode spacing value based on the working status of the array computing unit; In the case that the actual electrode spacing value changes suddenly, the actual electrode spacing value is incremented or decremented according to a preset step size.

9. The method for calculating the lateral electrode spacing of an array according to claim 1, wherein: The signal transmission efficiency and energy loss distribution of the array computing unit are obtained by: Multiple monitoring points are set in the array computing unit to collect the signal amplitude and energy loss value of each monitoring point in real time; Based on the collected signal amplitude and energy loss values, a signal transmission efficiency matrix and an energy loss distribution matrix are constructed; Through matrix operations, the overall signal transmission efficiency and energy loss distribution of the array computing units are calculated.

10. The method for calculating the lateral electrode spacing of an array according to claim 1, wherein: The adjustment process of the actual electrode spacing value includes: A plurality of adjustment modules are provided in the array computing unit, each adjustment module corresponding to an electrode spacing adjustment region; Based on the actual electrode spacing value, the electrode spacing of each adjustment module is adjusted respectively; During the adjustment process, the piecewise linear interpolation method is used to smooth the electrode spacing value.

Images