Delhi Jal Board (DJB), the public water utility serving Delhi, the capital city of India, was facing challenges in reducing pollution load through normal dosing systems. The traditional dosing system was not effective in reducing the levels of BOD and TSS to the desired minimum levels. Additionally, there was no real-time monitoring system in place to track the health of the plant and its equipment. DJB needed an AI-based solution that could automate and optimize the dosing process, and provide real-time monitoring and insights to improve the overall plant health and efficiency integrated with predictive maintenance, and AI-based calculations for better pollution load reduction.

Problem Statement

Our company Plaxonic worked with DJB to create an Intelligent Self-Administered and Self-Monitored Automatic Chemical (ISASMAC) dosing system. This system was engineered with AI/ML technology to not only adjust the dosing of coagulant with the flow, but also predict maintenance requirements. Our solution also integrated IoT devices to collect real-time data from the dosing system and other plant equipment, and provided an intuitive dashboard to monitor the system's health and utilization.

How Did We Help?

Our team followed a collaborative approach with DJB to understand their requirements and challenges. We then designed an end-to-end solution that included hardware, software, and IoT integration. We used agile methodologies to develop and deploy the solution in DJB plants.
&nbsp;
We developed a software system that would connect the dosing system along with the health information of all the electrical &amp; mechanical devices at a common platform. The software system was customized and configured as per the requirements of DJB and incorporated features for its integration with the relevant modules of at the plants and of laboratories such as OIML, LIMS, etc.
&nbsp;
The software system has the following features:
&nbsp;
<ul>
<li>IoT device integration for real-time monitoring of dosing system data, including health condition of E&amp;M devices, asset knowledge, and AI-based-analytics.</li>
</ul>
&nbsp;
<ul>
<li>Monitoring system units, health and utilization of the dosing system and critical assets with an intuitive, color-coded dashboard.</li>
</ul>
&nbsp;
<ul>
<li>Historical plant health and system utilization charts to provide insights into how the system is performing over time.</li>
</ul>
&nbsp;
<ul>
<li>Calculation of the reduction in pollution load after intervention of dosing and oxygen requirements in the aeration tank to neutralize the pollution load.</li>
</ul>
&nbsp;
<ul>
<li>Energy Savings Calculator for calculating the electrical savings.</li>
</ul>
&nbsp;
<ul>
<li>Fetching operational parameters like SVI, Sludge age, etc., on the basis of DJB&rsquo;s Laboratory analysis and suggesting measures for better treatment required after dosing.</li>
</ul>
&nbsp;
<ul>
<li>AI-based dosing pattern suggestions for better pollution load reduction.</li>
</ul>
&nbsp;
<ul>
<li>Preventive maintenance plan with records of preventive maintenance and repairs done and advanced alarm system.</li>
</ul>

Our Approach

The smart dosing system developed by Plaxonic utilizes the following technologies:
&nbsp;
<ul>
<li>Internet of Things (IoT)&nbsp;- To connect and communicate with various devices in the dosing system and to enable real-time monitoring and data sharing.</li>
</ul>
&nbsp;
<ul>
<li>Machine Learning (ML)&nbsp;- To analyze the dosing data and predict the best dosing patterns and also predict the maintenance required in the system.</li>
</ul>
&nbsp;
<ul>
<li>Big Data &amp; Analytics&nbsp;- To handle large volumes of dosing data, perform data analysis, and generate insights that can be used to optimize the dosing system and reduce pollution load.</li>
</ul>
&nbsp;
<ul>
<li>Artificial Intelligence (AI)&nbsp;- To enable AI-driven health scoring and report generation, as well as AI-based calculations for better pollution load reduction.</li>
</ul>
&nbsp;
The tech stack used for the development of the smart dosing system and software includes:
&nbsp;
<ul>
<li>Web -&nbsp;Next.js and Tailwind CSS for developing the web application</li>
</ul>
&nbsp;
<ul>
<li>API -&nbsp;Node.js for building the API layer, MongoDB for database management, and Apache ZooKeeper for distributed coordination and configuration management</li>
</ul>
&nbsp;
<ul>
<li>IoT controller -&nbsp;MQTT for controlling the IoT devices in the system.</li>
</ul>
&nbsp;
<ul>
<li>Database -&nbsp;MongoDB and InfluxDB for storing and managing the system and sensor data. Redis for caching.</li>
</ul>
&nbsp;
<ul>
<li>Data Analytics -&nbsp;Kafka, Python</li>
</ul>
&nbsp;
<ul>
<li>DevOps -&nbsp;Kubernetes for container orchestration and GitLab CI/CD for continuous integration and delivery.</li>
</ul>
&nbsp;
<ul>
<li>Project&nbsp;Management -&nbsp;Click-up</li>
</ul>

Tools & Technologies  

After four months of implementation, the system was successfully able to reduce the BOD and TSS levels to well below the targeted values of 10 PPM. The system recorded an average of 5 PPM BOD and 8 PPM TSS levels. With real-time monitoring and predictive maintenance, the system was able to improve the overall plant health and efficiency. DJB was highly satisfied with the outcome and plans to deploy the solution in more of its plants in the future.

The Result 

Continuous Growth, Evolving Technologies & Customer Success

customer success

AI Driven Smart Dosing System for a Public Utility Organization

Do you need help with something? We are here to assist you.

Quick Links

Services

Resources

Careers