Distributed AI Computation Units for Water Treatment Control

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional AI-based water treatment control systems do not effectively consider the installation position of AI, leading to inefficiencies in controlling multiple devices within a water treatment plant.

Innovation Solution

A water treatment plant architecture that includes a central monitoring device and multiple control devices, each equipped with a computation unit using a machine learning-based calculation model, allowing for localized AI processing near the water treatment apparatuses to reduce data transmission delays and enhance control processing efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If AI computation is centralized in the central monitoring device, then system structure is simplified, but data transmission delays increase and control processing efficiency decreases

Engineering Contradiction:

Improvesystem structureVSAvoiddata transmission delay

Core Design Contradiction:

Device complexityVSLoss of time

Solution Approach 1:

The patent divides the AI computation function into separate computation units that are distributed to individual control devices, rather than centralizing it in the central monitoring device. This segmentation allows each control device to independently perform AI-based computations locally, eliminating data transmission delays while maintaining a manageable system structure through modular architecture.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent implements local AI computation capabilities at each control device level, allowing computations to be performed where the data is generated and where control decisions are made. This local processing approach eliminates the need for data to be transmitted to and from a centralized AI processing location, thereby reducing transmission delays while maintaining system simplicity through consistent local architecture.

Inventive Principle:

Principle #3Local quality

2Device complexity

If AI computation is centralized in the central monitoring device, then system structure is simplified, but control processing speed decreases

Engineering Contradiction:

Improvesystem structureVSAvoidcontrol processing speed

Core Design Contradiction:

Device complexityVSProductivity

Solution Approach 1:

The patent segments the control processing function by distributing AI computation units to multiple control devices, enabling parallel processing of control tasks. This allows simultaneous execution of AI-based control algorithms across different water treatment apparatuses, significantly increasing overall control processing speed while maintaining a simplified centralized monitoring structure.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent transitions from a single-dimensional centralized processing model to a multi-dimensional distributed processing architecture. By adding the spatial dimension of distribution across multiple control devices while maintaining vertical integration through the central monitoring device, the system achieves both structural simplicity and enhanced processing throughput through parallel operations.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

3Loss of time

If multiple control devices with local computation units are deployed, then data transmission delays are reduced, but device complexity increases

Engineering Contradiction:

Improvedata transmission delayVSAvoidsystem structure

Core Design Contradiction:

Loss of timeVSDevice complexity

Solution Approach 1:

The patent implements a standardized computation unit design that can be universally deployed across multiple control devices. Each control device incorporates the same AI computation capabilities, allowing them to independently perform local processing without requiring different configurations. This universal approach reduces data transmission delays while preventing system complexity from increasing, as the same modular component is reused throughout the system.

Inventive Principle:

Principle #6Universality (Multi-functionality)

4Productivity

If multiple control devices with local computation units are deployed, then control processing efficiency is improved, but device complexity increases

Engineering Contradiction:

Improvecontrol processing efficiencyVSAvoidsystem structure

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent segments control processing into independent, self-contained computation units that can operate autonomously at each control device. This segmentation enables parallel execution of control algorithms across multiple apparatuses, dramatically improving overall processing efficiency. The modular nature of these segmented units maintains manageable complexity by allowing independent deployment and operation of each processing node.

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS11649183B2Water treatment plant

Publication Date: 2023.05.16 MITSUBISHI ELECTRIC CORP

AI summary

A water treatment plant includes a central monitoring device, a control device, a control device, and a computation unit, and causes a water treatment apparatus and a water treatment apparatus to execute water treatment. The central monitoring device monitors the water treatment apparatus and the water treatment apparatus. The control device performs a first control for the water treatment apparatus. The control device performs a second control for the water treatment apparatus. The computation unit is located outside the central monitoring device and performs a first computation related to the first control using a first calculation model generated by a first machine learning.